Skip to main content

Full text of "The ornithology of the Straits of Gibraltar"

See other formats


BRALTAR 


/HOWARD  IRBY,  F.L.S. 


BIOLOGY 
LIBRARY 


THE    ATHENJEUM 


N°  3595,  SEPT.  19,  '96 
SCIENCE 


The   Ornithology  of   the  Straits  of  Gibraltar. 

By    Lieut.  -  Col.    L.  Howard   L.    Irby. 

Second  Edition.  (Porter.) 
UNDER  the  above  title,  rather  more  than 
twenty  years  ago,  Col.  Irby  produced  a 
modest  little  octavo  volume  which  at  once 
secured  a  high  position  in  ornithological 
literature.  It  treated  of  the  natural  history 
of  the  north  -  western  portion  of  Morocco, 
from  Tangier  southwards  to  a  little  beyond 
Larache,  and  for  about  twelve  miles  inland, 
beyond  which  excursions  were  in  those  days 
difficult,  not  to  say  dangerous  ;  while  on  the 
Spanish  side  the  area  embraced  was  far 
wider,  comprising  the  country  between 
Cadiz,  Gibraltar,  Malaga,  Granada,  Cordova, 
Seville,  and  down  the  Guadalquivir  to  the 
great  marshy  delta  of  that  river.  Birds, 
their  habits  and  nidification,  with  the  dates 
of  arrival  and  departure  in  the  case  of 
migrants  —  the  results  of  several  years' 
experience — formed  the  principal  feature; 
but  there  was  a  great  deal  of  valuable 
information  about  mammals,  reptiles,  and 
other  subjects  in  the  work,  and  it  became  a 
handbook  for  the  naturalist  traveller  in  the 
South  as  much  as  the  book  of  reference  for 
the  student  at  home.  Almost  coincidently 
with  its  publication  Col.  Irby  retired  from 
the  74th  Highlanders,  but  he  by  no  means 
forgot  the  fascinating  country  with  which 
his  name  will  always  be  identified,  and  he 
subsequently  devoted  considerable  portions 
of  the  next  twenty  years  to  revisiting  his 
former  haunts  as  well  as  exploring  new 
districts.  In  doing  this  he  consorted  with, 
and  availed  himself  of  the  experience  of, 
many  kindred  spirits,  such  as  Major  Wil- 
loughby  Yerner  (one  of  the  most  daring  of 


cragsmen),  Capt.  E.  Savile  Eeid,  late  E.E., 
Mr.  A.  C.  Stark,  and"  others,  wnose  notes  are 
now  added  to  his  own.  The  result  is  that, 
instead  of  the  little  octavo  of  1875,  we  have 
a  handsome  quarto,  with  beautiful  coloured 
jpjates  in_Mr_A.  Thorburn's  best  style  from 
birds  living  in  Lord  LilforaTs'^avTaries'near 
Oundle,  as  well  as  numerous  illustrations 
of  scenery  from  sketches  and  photographs. 
Much  new  and  valuable  information  has 
been  added,  and  the  list  of  birds  is  raised 
from  335  to  369,  while  a  feature  of  the  pre- 
sent volume  is  the  short  diagnosis  of  each 
species  on  the  lines  adopted  by  the  author 
in  his  useful  •  Key  to  British  Birds,'  which 
was  noticed  in  these  columns  some  time 
ago.  Two  maps  and,  it  is  almost  unneces- 
sary to  say,  an  adequate  index  are  to  be 
found  in  both  editions. 

Owing  to  its  position  in  close  proximity 
\  to  Africa,  Spain — or  let  us  say  the  Iberian 
~  Peninsula,  to  avoid  offence  to  our  Portuguese 
friends — is  probably  the  most  interesting 
country  in  Europe  as  regards  its  fauna.  In 
the  south,  which  immediately  concerns  us, 
there  is  an  ichneumon  (Herpestes)  almost,  if 
not  absolutely,  identical  with  the  Numidian 
species  ;  the  lynx  is  peculiar,  and  has  just 
a  flavour  of  the  African  serval ;  the  red  deer 
of  the  north  comes  down  to  the  pine- clad 
shores  of  the  Mediterranean,  while  the  ibex 
from  the  mountain  ledges  looks  down  upon 
its  blue  waters — and  there  they  stop  ;  there 
is  no  sign  of  them  in  North  Africa.  As  for 
the  bear,  there  is  no  trace  of  it  in  historic 
times  to  the  south  of  Madrid,  while  the 
chamois  (izard)  is  not  found  below  the 
Pyrenees  and  the  Cantabrian  range ;  but 
otherwise  the  main  continental  fauna  is  well 
represented.  Even  among  birds,  in  spite  of 
their  powers  of  flight,  there  are  some  re- 
markable features  and  irregularities  in  dis- 
tribution. There  is  an  indigenous  Iberian 
short-toed  lark,  quite  different  from  the 
ordinary  species  of  the  rest  of  Europe,  though 
barely  distinguishable  from  an  inhabitant 
of  North  Africa  ;  there  is  a  larger  lark  with 
a  decurved  bill  belonging  to  an  African, 
and  almost  a  Saharan,  group ;  the  Iberian 
green  woodpecker  differs  from,  that  of 
all  the  rest  of  Europe ;  and  there  is 
a  fine  white-shouldered  eagle,  the  repre- 
sentative of  the  Eastern  imperial  eagle, 
which  is  barely  known  this  side  of 
Italy.  For  the  nearest  relative  of  the 
beautiful  Iberian  azure  -  winged  magpie 
( Cyanopica  cooki}  all  Europe  and  the  greater 
part  of  Asia  must  be  crossed,  until  South- 
Eastern  Siberia  and  China  are  reached. 
Our  familiar  pie  is  found  throughout  Spain 
down  to  the  woods  of  the  Guadalquivir; 
but  in  the  mountainous  districts  of  the 
Alpuj  arras  and  on  the  skirts  of  the  Sierra 
Nevada  the  magpies  become  dark,  and 
exhibit  decided  affinities  with  a  North 
African  species.  The  Spanish  grey  shrike 
is  virtually  confined  to  the  Peninsula,  where 
its  range  extends  to  the  extreme  southern 
shores  ;  yet  the  bird  never  crosses  the  in- 
tervening ten  or  twelve  miles  to  Africa,  nor 
does  its  representative  there — a  species 
which  only  a  naturalist  can  distinguish — 


ever  invade  Spain.  An  orange  -  eating 
bulbul  (Pycnonotus)  is  found  at  Tangier, 
but  never  crosses  the  Straits  to  the  land 
which  is  said  to  be  "famous  for  oranges 
and  pretty  women."  The  buff  -  backed 
heron  (the  representative  of  the  Eastern 

"paddy-bird")  breeds  freely  in  the  south 
of  Spain,  but  the  rest  of  Europe  hardly 
knows  it,'  even  in  swamps  apparently 
the  most  suitable,  and  to  find  it  again  we 
must  visit  North  Africa.  The  little  three- 
toed  quail  is  another  African  species  found 
breeding  locally  in  the  palmetto  (Chamserops) 
scrub  of  Andalucia,  and  also  near  Grirgenti, 
where  Sicily  is  almost  in  sight  of  Tunisia  ; 
nowhere  else  in  Europe.  These  are  only  a 
few  of  the  points  which  occur  to  us,  and  it 
may  be  imagined  that  the  natural  history 
of  a  country  which  presents  these  paradoxes 
is  worth  writing  about. 

In  a  book  which  deals  with  facts  set  down 
in  the  most  straightforward  manner  it  is 
not  easy  to  find  passages  suitable  for  quota- 
tion, for  even  Major  Verner's  feats  on  the 
cliffs  and  his  narrow  escapes  when  visiting 
the  nests  of  eagles  and  vultures  are  not 
fully  described  here.  The  nests  were  in 
such  and  such  a  place,  of  such  materials, 
and  contained  so  many  eggs  on  such  a 
date ;  but  there  is  nothing  to  make  your 
flesh  creep.  We  notice  that  the  experiences 
of  all  the  contributors  to  this  work  are  to 
the  effect  that  the  bearded  vulture — often 
called  Liimmergeier — is  simply  a  scavenger, 
and  never  attempts  to  touch  kids,  lambs,  or 
indeed  any  living  mammal.  It  is  much  if 
the  maligned  bird  occasionally  takes  a  snake 
or  a  lizard,  though  it  certainly  does  pick  up 
small  tortoises  (as  it  does  bones)  and  drop 
them  on  the  rocks  to  break  them :  a  per- 
formance which  is  said  to  have  caused  the 
death  of  /Eschylus,  and  the  tradition  of 
which  once  saved  the  life  of  a  bearded 
vulture.  Turning  to  smaller  birds,  there 
is  a  novel  and  interesting  description 
of  the  nesting  habits  of  the  handsome 
black  wheatear,  a  lively  sooty  bird  with  a 
white  rump,  which  is  locally  known  as 
pedrero,  from  its  habit  of  building  founda- 
tions of  small  stones,  and  even  a  wall  or 
screen  in  front  of  its  nest.  One  of  these 
barriers — of  which  an  illustration  is  given — 
was  9  in.  long,  the  same  in  width,  2£  in. 
high,  and  consisted  of  282  stones,  while 
the  foundation  contained  76  more,  the 
total  weight  of  the  stones  being  4£lb. 
Then  there  are  some  valuable  bits  of  in- 
formation respecting  the  names  of  birds, 
illustrating  Arab  superstitions,  as  in  the 
passage  referring  to  the  white  owl — which 
is  too  long  to  quote — and  that  on  the  lap- 
wing. We  always  wondered  why  the  latter 
bore  the  trivial  name  Judia  in  Spain 
(El  Thudi  in  Morocco),  but  it  appears  that 
"  the  superstitious  Arabs  believe  that  these 
birds  are  Jews  changed  into  the  shape  of 
birds,  and  also  believe  that  they  retain  all 
their  Israelitish  characteristics,  even  wear- 
ing the  black  Hebrew  skull-cap."  And 
with  this  we  close  our  notice,  having,  we 
trust,  said  enough  to  show  that  a  book, 
although  technical,  is  not  necessarily  dry. 


BEARDED    VULTURE. 
Gypaetus  barbatus.   ad. 


Litho.  W.  Greve,  Berlin. 


THE 


ORNITHOLOGY 


OF    THE 


STBAITS    OF    GIBRALTAR 


BY 

LIEUT.-COLONEL  L.  HOWARD  L.  IRBY,  F.L.S., 

// 

LATE    SF.VENTY-FOt'RTH    HIGHLANDERS. 


SECOND    EDITION, 

REVISED  AND  ENLARGED. 


WITH     AN    APPENDIX 

COKTAIKIKG  A  LIST  OF  THE  LEPIDOPTEEA  OF  THE  NEIGHBORHOOD. 


FLUMINA     AMO     SYLVASQUE     INGLORIUS. 


LONDON: 

R.    H.    PORTER, 

18    PRINCES    STREET,  CAVENDISH    SQUARE. 

1895. 


PRINTED    BY    TAYLOR    AND    FRANCIS, 

RED  LION  COURT,  FLEET   STREET. 


CONTENTS. 


Page 

LIST  OF  ILLUSTRATIONS v 

INTRODUCTION 

APPENDIX 311 

INDEX  TO  BIRDS 


811998 


LIST    OF     ILLUSTRATIONS. 


FULL-PAGE  PLATES. 

Page 

Bearded  Vulture.     Thorburn     Frontispiece 

Blue-winged  Magpie.     Thorburn    To  face  83 

Black  or  Cinereous  *  Vulture.     Thorburn     „  146 

Golden  Eagle.     Thorburn     „  171 

White-shouldered  Eagle.     Thorburn     *         „  173 

Descent  to  Nest  of  Bonelli's  Eagle.     Smit,  after  Verner •  •  175 

Interior  of  Cavern  with  Nest  of  Bonelli's  Eagle.     Smit,  after  Verner    ....         „  177 

Booted  Eagle.     Thorburn     „  181 

Mediterranean  Peregrine.     Thorburn     „  191 

The  First  Rise  of  the  Grey-lags,  Laguna  de  la  Janda.      Smit,  after 

Verner    „  217 

European  Bush-Quail.     Thorburn „    .  241 

Bustard-Driving.     Smit,  after  Verner     „  255 

A  Deep  Place  in  the  Malabrigo  Soto.     Smit,  after  W.  F. ,  279 

Whiskered  Terns  at  Home.     Smit,  after  Verner  „  293 


ILLTTSTHATIONS  PRINTED  IN  THE  TEXT. 


NOTE 


THE  Illustrations  printed  in  the  text  (excepting  the  two 
diagrams  of  Bird)  are  from  photographs  taken  by  Major 
Willoughby  Verner,  Rifle  Brigade,  of  places  visited  by  the 
Author  in  1894. 


,..  r  >  HTTTne  enu  or  voitnne. 

Map  of  Morocco \ 


*  Misspelt  "  Ciiierous  ''  on  Plate. 


LIST    OF     ILLUSTRATIONS. 


FULL-PAGE  PLATES. 

Page 

Bearded  Vulture.     Thorburn     Frontispiece 

Blue-winged  Magpie.     Thorburn    To  face  83 

Black  or  Cinereous  *  Vulture.     Thorburn     „  145 

Golden  Eagle.     Thorburn     „  171 

White-shouldered  Eagle.     Thorburn     ^  „  173 

Descent  to  Nest  of  Bonelli's  Eagle.     Smit,  after  Verner „  175 

Interior  of  Cavern  with  Nest  of  Bonelli's  Eagle.     Smit,  after  Verner    ....  „  177 

Booted  Eagle.     Thorburn     „  181 

Mediterranean  Peregrine.     Thorburn     „  191 

The  First  Kise  of  the   Grey-lags,  Laguna  de  la  Janda.       Smit,  after 

Verner    „  217 

European  Bush-Quail.     Thorburn „  241 

Bustard-Driving.     Smit,  after  Verner     „  255 

A  Deep  Place  in  the  Malabrigo  Soto.     Smit,  after  W.  F.  „  279 

Whiskered  Terns  at  Home.     Smit,  after  Verner   „  293 


ILLUSTEATIONS  PBINTED  IN  THE  TEXT. 

Diagram  of  Bird    vi 

Diagram  of  Inner  side  of  Wing  vii 

Terrace  on  the  north  side  of  San  Bartolome" 28 

El  Organo,  near  Tarifa 30 

Peasantry,  Tapatanilla 32 

Laja  del  Ciscar 33 

Nest  of  the  Black  Wheatear    42 

Nest  and  Egg  of  Griffon  148 

Cave  with  Griffons'  Nests 150 

Young  Griffon  in  Cavern 150 

Nest  of  Neophron,  showing  Crag    153 

„            „         nearer  View  153 

Nesting-place  of  Neophron   155 

Nest  of  Bearded  Vulture 157 

v           »            „        160 

„          „            „       near  View    160 

Nest  of  White-shouldered  Eagle 172 

Laja  de  la  Zarga,  Nesting-place  of  Bonelli's  Eagle  and  Griffons    176 

Nest  of  Snake-Eagle  in  Cork-tree   183 

Another  Nest  of  Snake-Eagle  in  Cork-tree    183 

Laja  del  Ciscar 310 


Map  of  South-western  Andalucia     1  at  the  end  of  Volume. 

Map  of  Morocco  .  , 


Misspelt  "  Cinerous  "  on  Plate. 


Lores  Ear-fouerts 

\    Forehead     ', 


\     roreneaa      »     f  "*• 

\     I  ...I./  .""" 

v-j-evow/.^/^--..^ 

\i— *~~~~>/'          "x 


..Nape 


.Upper  Back 


Lower  Bach 


Diagram,     of     a     Bird, 

TO  ILLUSTRATE  THE  TEKMINOLOGY  OF  THE  PLUMAGE. 


ORNITHOLOGY 


OF  THE 


STRAITS    OF    GIBRALTAR. 


INTRODUCTION. 

THE  list  of  birds  and  ornithological  notes  contained  in  this 
book  are  compiled  from  observations  made  on  both  sides  of 
the  Straits  of  Gibraltar — on  the  African  side  within  a  region 
extending  from  Tangier  southwards  to  the  lakes  of  Has  el 
Doura,  beyond  Larache,  and  eastward  from  Tangier  to  Tetuan 
and  Ceuta,  not  reaching  inland  more  than  ten  or  twelve  miles ; 
on  the  European  side  is  included  that  part  of  Andalucia  which 
would  be  bounded  by  an  imaginary  line  drawn  from  Gibraltar 
to  Malaga,  thence  to  Granada,  Cordova,  and  Seville,  along  the 
delta  of  the  Guadalquivir  to  Cadiz. 

Nearly  all  the  information  relating  to  the  birds  of  the  Spanish 
side  of  the  Straits  is  collected  from  personal  observations  made 
during  a  more  or  less  prolonged  stay  at  the  Rock,  between 
February  1868  and  May  1872,  and  again  from  February  to 
May  1874,  but  including  during  this  time  only  one  summer 
period,  viz.  July,  August,  and  the  first  half  of  September. 
For  the  first  three  years  of  my  residence  at  Gibraltar  I  was 
quartered  with  my  regiment,  the  remaining  time  being  passed 

B 


2  INTRODUCTION. 

there  chiefly  with  a  view  to  ornithological  pursuits,  from  time 
to  time  making  excursions,  generally  of  about  a  fortnight's 
duration,  to  some  part  or  other  within  the  districts  above 
mentioned,  but  chiefly  confining  my  attentions  jto  the  country 
within  a  day's  journey  of  Gibraltar. 

The  observations  on  the  Moorish  birds  are  in  a  great  measure 
culled  from  the  MS.  of  the  late  M.  F.  Favier,  a  French  collector 
well  known  to  the  ornithological  world,  who,  after  a  residence 

**  *     * 

•£•„•  of  about  thirty-one  years  at  Tangier,  died  there  in  1867.  I 
";  was  informed  that  he  had  left  a  MS.  written  in  French, 
containing  his  notes  on  birds ;  but  was  not  permitted  by  the 
owner  to  do  more  than  glance  at  it,  although  he  offered  it 
for  sale  at  what  seemed  to  me  a  very  exorbitant  price.  Fearing 
to  lose  a  book  which  might  prove  of  considerable  utility  in 
the  pursuit  of  my  favourite  science,  I  yielded  to  his  demands 
and  secured  the  coveted  prize,  but  found  upon  perusal,  amidst 
a  mass  of  bad  grammar,  bad  spelling,  and  worse  writing,  which 
cost  many  hours  to  decipher,  that  it  did  not  contain  so  much 
information  as  there  was  reason  to  anticipate,  a  good  deal  of 
the  matter  having  been  copied  from  other  authors. 

However,  there  was  some  grain  among  all  this  chaff;  and 
such  facts  and  details  as  are  considered  worth  recording  are 
given  below  with  Favier's  name  attached,  and  supplemented 
by  my  own  observations  in  Morocco.  These  notes  are  kept 
separate  from  those  referring  to  the  Spanish  side  of  the  Straits. 

This  little  work  is,  it  may  be  distinctly  understood,  not 
intended  to  give  any  special  information  to  scientific  ornitho- 
logists, but  is  published  with  the  view  of  assisting  with  trust- 
worthy information  any  amateur  collectors  who  visit  South- 
western Europe ;  and  it  is  hoped  especially  that  it  may  be 
useful  to  officers  who,  like  the  \vriter,  may  find  themselves 
quartered  at  Gibraltar.  For  it  admits  of  little  doubt  that  the 
study  of  Natural  History  will  always  help  to  pass  away  with 
pleasure  many  hours  that  would  otherwise  be  weary  and  tedious 


LITERATUKE.  3 

during  the  time  military  men  may  have  to  "  put  in "  at  dear, 
scorching  old  "  Gib." 

There  is  ample  room,  for  any  one  with  energy,  to  work  out 
a  great  deal  more  information  on  the  birds  of  the  Straits ;  but 
it  must  be  remembered  that  little  can  be  done  in  hasty  visits 
of  two  or  three  months,  or  by  comparing  skins  secured  by 
assistants,  called  "  collectors,"  who  know  nothing  of  the  habits 
of  the  birds  they  send  to  those  who  employ  them,  and  upon  whose 
veracity  even  as  to  locality  the  employer  cannot  implicitly  depend. 

It  may  be  as  well  to  notice  such  papers  &c.  as  have  been 
published  hitherto  relating  to  the  ornithology  of  the  district 
of  the  Straits.  Dismissing  the  Spanish  lists  as  meagre  and 
full  of  errors,  we  commence  with  the  papers  written  by  Lord 
Lilford  in  'The  Ibis'  for  1865  (p.  166)  and  1866  (pp.  173 
&  377).  In  addition  to  the  interesting  facts  they  contain,  these 
essays  are  the  first  which  give  any  reliable  information  on  the 
subject,  and  lay,  as  it  were,  the  foundation  of  all  the  work  that 
has  since  been  done  with  regard  to  Spanish  ornithology. 

Since  then  Mr.  Howard  Saunders  has  written,  also  in  '  The 
Ibis'  (1871,  pp.  54,  205,  &  384),  a  "  List  of  the  Birds  of 
Southern  Spain,"  extending  as  far  eastward  as  Minorca  and 
as  far  to  the  north  as  the  fortieth  degree  of  latitude,  thus 
comprising  a  very  large  area.  He  has  also  contributed  some 
other  papers  to  'The  Ibis'  (1869,  pp.  170  &  391),  which, 
altered  and  considerably  enlarged,  have  appeared  in  '  The 
Field '  under  the  head  of  "  Ornithological  Kambles  in  Southern 
Spain." 

On  the  African  side,  Mr.  G.  W.  H.  Drummond  Hay 
(Proceedings  Zool.  Society,  1840,  p.  133)  gave  a  list  of  birds 
noticed  by  him  at  Tangier,  but  only  fifty-five  in  number,  while 
Mr.  Tyrwhitt-Drake  wrote  a  list  of  the  birds  observed  by  him 
in  Tangier  and  Eastern  Morocco  (Ibis,  1867,  p.  421) ;  and  a 
"  List  of  Birds  seen  near  Tangier "  appeared  in  '  Naumannia, 
but  only  a  list,  and  not  a  very  accurate  one. 

B2 


4  INTRODUCTION. 

Lists  of  birds,  generally  speaking,  have  very  few  dates 
appended;  the  exact  localities  where  a  species  may  be  found 
are  seldom  indicated ;  the  period.- of  migration  is  also  not  often 
stated.  "Breeds  plentifully,"  "  appears  in  winter,"  "a  regular 
visitant,"  "  abundant  in  spring," — such  are  usually  the  vague 
remarks  given  with  each  species. 

Many  of  the  ornithological  papers  in  periodicals  and  journals 
are  written  up  from  one  or  two  passing  visits,  often  very 
short  ones ;  and  some  of  the  writers  possess  a  power  of  vision 
truly  astonishing.  They  see  a  bird  in  the  distance,  imagine 
it  to  belong  to  a  certain  species,  at  last  believe  it  to  be 
such,  and  end  by  placing  the  name  in  their  note-book,  to  appear 
in  due  time  in  print. 

The  only  way  to  avoid  such  errors  is  never  to  include  any 
bird  in  a  list  except  when  actually  obtained  and  identified. 
It  often  happens,  also,  that  the  bird  seen  and  included  is  one 
which  it  would  be  quite  impossible  to  distinguish  from  another 
closely  allied  species  without  handling  them  both. 

These  remarks  may,  no  doubt,  appear  very  invidious ;  but 
it  is  want  of  accuracy  in  such  matters  which  renders  utterly 
futile  any  attempt  to  make  out  the  distribution  of  birds. 

Local  names,  often  trivial  and  unimportant,  must  generally 
be  accepted  cum  grano  salis ;  for,  unless  long  resident  and 
conversant  with  the  language  of  the  country,  the  compiler  is 
apt  to  fall  into  the  same  class  of  errors  as  those  of  the  celebrated 
Count  Smorltork,  who  would  probably  have  written  the  English 
name  of  the  Curlew  according  to  the  story  told  of  a  gunner 
in  the  Eastern  Counties,  who,  when  asked  by  a  portly  old 
citizen,  "  What  do  you  call  those  birds  I  "  replied,  "  Bless  you, 
Curlews  we  generally  calls  'em ;  but  when  we're  vexed  with 
'em,  we  calls  'em  beggars."  These  vernacular  names  are  most 
useful,  of  course,  in  the  case  of  the  more  common  species, 
and  in  Andalucia  are,  in  many  instances,  of  Arabic  derivation, 
relics  of  the  Moorish  occupation  and  of  days  when  under  their 


VERNACULAR  NAMES.  5 

rule  Spain  was  flourishing,  when  all  that  is  'worth  seeing 
was  built,  all  that  is  artificially  good  being  remnants  of  the 
work  of  the  then  industrious  Moors.  Where  are  the  latter  now 
as  a  nation  1 

As  a  proof  of  the  inaccuracy  of  local  nomenclature,  a  single 
name  is  often  applied  to  several  species,  sometimes  not  even 
belonging  to  the  same  genus.  Thus  Aguila,  Aguiluclio, 
according  to  the  ideas  of  the  individual,  may  be  any  of  the 
Diurnal  Accipitres,  from  a  Lammergeyer  to  a  Lesser  Kestrel ; 
and  they  are  even  occasionally  used  to  designate  the  Raven ! ! 

So  Bvjo  applies  to  all  Owls,  Culiblanco  to  all  Wheatears ; 
Chorlito,  the  real  name  of  the  Golden  Plover,  is  used  for 
various  Waders ;  while  Pitillo,  Frailecillo,  Andarios,  Correrios 
are  indefinite  names  applicable  to  any  small  Waders  and  some 
larger  ones.  Pito  real  near  Gibraltar  is  Picas  major,  our 
Great  Spotted  Woodpecker ;  near  Seville  it  is  Gecinus  Sharpti, 
the  representative  of  our  English  Green  Woodpecker  ( G.  viridis). 
Carpintero  in  Central  Spain,  according  to  Lord  Lilford,  is 
Picus  major  ;  near  Gibraltar  it  is  the  Great  Titmouse  (Parus 
major].  Lavandera,  or  "  washerwoman,"  according  to  localities 
is  either  a  Wagtail  or  a  Green  Sandpiper.  Quebrantahuesos, 
"  bone-breaker,"  properly  applies  to  the  Lammergeyer  ;  but 
where  that  species  is  absent  it  is  usurped  by  the  Neophron. 

These,  among  other  instances,  prove  local  names  to  be  only 
an  assistance,  and  not  always  to  be  taken  to  signify  the  bird 
to  which  they  are  affixed. 

On  the  other  hand,  some  names  are  distinctive,  as  Alejaruco, 
Bee-eater ;  Abubilla,  Hoopoe  ;  Abujeta,  Godwit ;  Alcaravan, 
Stone-Curlew,  &c. 

The  Moorish  Arabic  names  are  for  the  most  part  copied 
from  Favier's  MS. ;  but  none  are  quoted  unless  corroborated 
by  reference  to  natives  of  the  localities  in  which  the  birds 
were  shot.  It  may  be  further  added  that,  as  it  is  a  matter 
of  considerable  difficulty  to  express  Arabic  words  by  English 


6  .       INTRODUCTION. 

letters,  in   this   work  a  mere   approximation   to   the   sound   is 
attempted  to  be  given. 

In  the  interest  of  the  sportsman  and  the  amateur  collector 
of  specimens,  it  has  been  endeavoured  to  give  a  few  hints  as 
to  the  localities  where  each  may  best  gratify  his  tastes ;  but 
it  would  be  foreign  to  the  purpose  and  inapplicable  to  the 
limits  of  this  work  to  reproduce  any  of  the  varied  information 
which  is  to  be  found  in  the  pages  of  Ford  and  other  guide- 
books. In  the  country  districts  of  Spain,  and  especially 
throughout  Andalucia,  nothing  ever  changes  except  the 
Government.  The  country  is  still  the  Spain  of  Ford,  we  might 
almost  say  of  Don  Quixote,  and  will  probably  remain  so  for 
centuries,  except  as  regards  the  power  of  the  priesthood,  which 
is  gradually  waning  and  will  doubtless  soon  cease  to  exist. 

In  a  general  sense,  travelling  in  Morocco  is  attended  with 
more    expense   and    less   comfort   than   in   Spain.      The    total 
absence  of  inns  in  the  former  country  (except  one  or  two  at 
Tangier   and   a   few   coast-towns)  renders  it   necessary  for  the 
European  traveller  to  carry  about  not  only  a  tent  but  a  good 
deal   of  extra   baggage,  cooking-appliances,  &c.,  which   would 
otherwise   be   superfluous.      The   impedimenta   are   transported 
on  the  backs  of  mules,  which,  may  be  hired  at  the  rate  of  one 
dollar  per   diem.      One  or  two  Moors  must  also  be  taken  to 
pitch   tents,  load   and   unload  the  pack-animals,  and  so  make 
themselves    generally    useful,   which    they    always    do.      They 
were   usually  paid   one  and  sixpence   a   day  each.      The   only 
provisions  which  can  be  procured  at  the  country  villages  consist 
of  fowls,  eggs,  butter,  milk,  and  kuskoo-soo;    the  latter   is   a 
peculiar  preparation  of  flour,  and  may  be  considered  the  national 
dish  of  Morocco.     It  is  therefore  advisable  for  a  party  travelling 
in  the  interior  to  provide  themselves  with  some  tins  of  preserved 
meat   and   vegetables,   as  well  as  with  whatever   wine,   spirits, 
&c.  they  might  require.     The  most  satisfactory  way  of  making 
an  expedition  through  the  country,  I  have  found  by  experience, 


-    MOROCCO.  7 

is  to  contract  with  a  respectable  Moor,  who  will  -usually  defray 
the  entire  expenses,  including  hire  of  baggage,  animals,  servants, 
cook,  and  meals,  exclusive  of  wine,  at  the  rate  of  five  dollars 
a  head  per  diem. 

As  far  as  I  have  had  opportunities  of  judging,  I  have  reason 
to   consider  travelling  in  Morocco  to   be  unattended  with  any 
danger  near   the   coast,  but   not   east   of  Tetuan,    in   the   Riff 
country,  or  in  mountainous  districts ;    and  to  Englishmen  the 
natives  are  certainly  better  inclined  than  to  any  other  European 
nation.     The  late  Sultan,  however,  issued  an  edict  to  the  effect 
that    he   would   not   hold   himself  responsible   for   the   life   or 
property  of  any  Christian  who  at  the  time  of  any  outrage  should 
be  unattended  by  a  Moorish  soldier ;  and  a  mounted  soldier  to 
act   as   guard   can   always   be   procured  on  application  to   the 
Consul   of  the  nation  to  which   the   applicant   belongs.      Any 
party   intending  to   make   an   expedition   further   than   ten   or 
fifteen  miles  from  Tangier  should  provide  themselves  with  this 
necessary  functionary,  to  whom  they  must  pay  one  dollar  a  day. 
This    sum   is    generally   considered   to   be   exorbitant  ;    and    it 
certainly  is  so  in  a  country  where  the  necessaries  of  life  are 
far  cheaper  than  in  any  country  in  Europe.     The  real  truth, 
however,  is  that   the  Moorish  authorities  throw  as  many  diffi- 
culties as  they  possibly  can  in  the  way  of  inquisitive  European 
travellers,    well   knowing    that,   were    the   interior   thoroughly 
opened  up,  the  vile  system  of  oppression  and  misgovernment  to 
which  it  is  and  has  been  for  so  long  a  victim  would  vanish 
before  the  opinion  of  the  civilized  world. 

The  vicinity  of  Tangier  is  as  good  a  ground  for  the  ornitho- 
logist as  can  be  wished  anywhere ;  but  it  has  been  a  great  deal 
worked  up  by  Olcese,  who  succeeded  Favier  as  the  naturalist  of 
Tangier.  About  twelve  miles  to  the  south  are  the  lakes  of  Sharf 
el  Akab,  well  worth  visiting  for  aquatic  birds.  The  country 
beyond  this  to  Larache  is  not  good  until  within  the  neighbour- 
hood of  that  town,  where  there  is  plenty  of  both  marshy  and 


8  INTRODUCTION. 

dry  ground,  the  latter,  in  places,  better  wooded  than  usual. 
Near  Larache,  on  the  north  bank  of  the  river,  are  the  ruins  of 
the  ancient  Lixus,  at  or  near  the  spot  where  Hercules  is  supposed 
to  have  conquered  Antaeus,  the  founder  of  Tangier,  which  takes 
its  name  from  his  wife  Tinga. 

South  of  Larache  are  the  lakes  of  Meshree  el  Haddar  (the 
talkers'  ford)  and  Has  el  Doura,  the  latter  running  for  miles 
southwards  in  the  direction  of  Rabat.  These  lakes  swarm  with 
every  kind  of  aquatic  bird,  according  to  the  season ;  but  in  the 
breeding-time  the  mosquitoes  are  enough  to  drive  any  European 
away,  besides  which  the  nests  are  so  plundered  by  the  Arabs 
that  it  is  hardly  worth  while  going  there  for  them.  Further 
south  than  this  I  have  not  been,  and  refer  my  readers  for  any 
information  to  Mr.  Drake's  paper  in  '  The  Ibis '  (I.  c.). 

Eastward  of  Tangier,  taking  the  road  to  Tetuan,  there  is  little 
or  nothing  to  be  done  in  the  way  of  birds  until  the  latter  place 
is  reached,  after  a  long  and  tedious  day's  journey ;  indeed  all 
that  part  of  Morocco  which  I  have  visited  is  very  wearisome  to 
travel  over,  except  near  Tetuan  and  Ceuta,  where  the  mountains 
break  the  sameness  of  the  route,  and  where  alone  any  true 
beauty  of  scenery  is  to  be  found. 

Of  these  hills  only  those  in  the  immediate  vicinity  of  Tetuan 
can  be  visited,  owing  to  the  lawless  character  of  the  hill  tribes 
and  their  Mahometan  prejudices,  and,  last  but  not  least,  owing 
also  to  the  exaggerated  stories  made  up  to  prevent  any  European 
from  travelling  about.  In  a  stream  from  one  of  these  mountains, 
to  the  south  of  Tetuan,  a  species  of  trout  (Salmo  macrostigma]  is 
found ;  they  are  also  probably  met  with  in  other  places,  which 
are  forbidden  ground  to  the  European. 

The  country  about  Tetuan  is  alike  interesting  to  the  ornitho- 
logist and  favourable  to  the  sportsman ;  about  Martine  are 
some  fine  marshes,  while  beyond  Cape  Negro,  towards  Ceuta, 
is  a  large,  irregularly  shaped,  shallow  laguna,  called  Esmir,  with 
great  masses  of  rush  and  sedge  interspersed  with  tamarisk 


APES'  HILL.  9 

bushes,  separated  from  the  sea  by  a  wide  sandbank  covered 
with  brushwood ;  this  laguna  and  marshes  are  by  far  the  best 
ground  to  be  seen  on  either  side  of  the  Straits.  Proceeding 
from  Esmir,  towards  Ceuta,  the  road  lies  either  on  the  shore  or 
along  the  usual  scrub-covered  country  till,  turning  to  the  left  by 
some  Roman  ruins,  a  pass  leading  up  to  the  Sierra  Bullones  is 
entered,  when  the  scenery  becomes  very  fine,  the  track  ascending 
by  the  side  of  a  bright  clear  stream,  through  bushes  sometimes 
so  thick  as  to  completely  shut  out  the  sky  overhead,  at  other 
times  passing  through  heather,  in  places  twenty  feet  high.  The 
path  becomes  gradually  worse,  till  the  climax  is  reached  in  the 
ascent  of  a  steep  hill  where  the  brushwood  tears  the  load  oif  the 
mules,  and  with  the  stones  and  rocks  nearly  renders  progress 
impossible.  Once,  however,  at  the  top,  a  fair  enough  road  is 
found  leading  to  the  village  of  Beut,  situated  in  a  sort  of  plateau 
at  an  elevation  of  about  1000  feet,  separated  from  Jebel  Musa 
by  a  deep  valley,  a  high  range  of  rocks,  and  another  shallow 
valley. 

About  here  nothing,  ornithologically  speaking,  is  to  be  seen 
(excepting  a  few  Choughs)  that  is  not  to  be  met  with  elsewhere. 
We  found  an  Eagle  nesting  on  the  north  face  of  the  range  south 
of  Jebel  Musa :  the  nest  was  in  a  most  difficult  position  to  get 
at ;  and  not  being  able  to  reach  any  place  near  enough  from 
which  to  shoot  the  Eagle,  we  left  the  eggs  as  worthless,  because 
unable  to  identify  the  bird  :  however  there  is  little  doubt  that 
it  was  the  Golden  Eagle.  We  saw  some  apes  about  the  rocks ; 
they  were  rather  wild,  and  lost  no  time  in  making  their  way  to 
the  top. 

The  view  from  this  sierra  (Apes'  Hill  of  the  English,  Jebel 
Musa  of  the  Moors,  Sierra  Bullones  of  the  Spaniards,  Abyla  of 
the  ancients)  is  magnificent,  and  baffles  description,  well  repay- 
ing all  the  trouble  and  difficulties  of  the  ascent. 

To  the  south  beyond  Tetuan  lie,  half  hidden  in  blue  mist, 
the  snow-streaked  mountains  of  the  Atlas,  stretching  far  away 


10  INTRODUCTION. 

out  of  sight,  the  summit  of  one  vast  snowy  pile  rather  to  the 
south-east  appearing  to  be  as  high,  and  looking  quite  as  white, 
as  the  Sierra  Nevada,  near  Granada,  which  is  also  distinctly 
visible  to  the  north-east;  but  this  African  snowy  range  seems 
further  off.  Below  to  the  east,  stretched  out  as  if  on  a  ground- 
plan  close  to  your  feet,  is  Ceuta,  with  its  ugly  white-washed 
Spanish  forts  and  towers,  between  which  and  the  Tetuan  river 
looms  the  gloomy  headland  of  Cape  Negro.  Westward  of  this 
are  range  after  range  of  comparatively  low  dark  hills,  rolling 
away  towards  Tangier  and  Cape  Spartel,  here  and  there  one  or 
two  being  topped  with  a  few  crags  and  rocks.  Far  to  the  west 
is  the  Atlantic,  leading  to  the  deep-blue  Straits,  looking,  as  they 
separate  Europe  and  Africa,  like  some  mountain-lake. 

Tarifa,  to  the  north-west,  is  clearly  visible,  as  are  the  crags 
of  the  Sierra  de  San  Bartolome,  the  sandy  cliffs  of  Cape 
Trafalgar,  and  the  long  spit  of  land  on  which  is  the  light- 
house ;  while  all  the  grey,  bare,  barren-looking  Spanish  sierras 
look,  with  the  sun  shining  on  them,  as  if  they  lay  within  a 
stone's  "throw. 

Gibraltar  was  shut  out  from  our  view,  owing  to  the  pleasantry 
of  some  Moors,  who  rolled  large  stones  down  the  only  path 
leading  to  the  summit  of  the  highest  peak,  and  so  prevented  us 
from  ascending.  However  the  view  we  did  get  was  such  as  can 
never  be  forgotten,  and  it  was  long  before  we  descended  to 
continue  our  hunt  for  birds. 

The  tops  of  these  mountains,  which  are  2600  feet  high  (the 
highest  part  of  JebelMusa  is  about  2800  feet),  were  covered 
with  thousands  of  violets  then  in  full  bloom.  The  flowers  were 
light-coloured  when  growing  among  the  stones  and  waterworn 
rocks  exposed  to  the  sun,  dark  when  shaded  and  growing  among 
the  stunted  bushes  which  were  scattered  about  here  and  there : 
their  scent  was  perfect.  Very  few  other  flowers  were  growing 
on  the  tops ;  but  most  conspicuous  among  them  was  the 
Gibraltar  candytuft ;  and  the  everlasting  palmetto  was  met  with 


LARGE  CAVE.  11 

at  the  very  highest  places.  The  base  of  all  "these"  crags  or  cliffs 
is  approached  by  a  steep  talus  of  small  broken  rocks,  among 
which  grows  a  very  thick  jungle  of  stunted  cork-  and  olive-trees 
about  15  feet  high. 

On  the  north  side  of  the  range  next  to  Beut  and  at  the 
western  end  of  it,  at  the  base  of  the  cliff,  is  a  wide  cave,  which, 
at  some  distance  from  the  entrance,  branches  off  into  two 
distinct  caves,  one  going  up  hill,  the  other  down.  My  com- 
panion ascended  the  upper  one  till  he  heard  in  the  darkness  the 
growling  of  some  beast,  probably  a  lynx  or  some  wild  cat ;  so  he 
returned,  and  we  collected  together  from  outside  a  lot  of  dead 
sticks  and  rubbish,  which  we  dragged  up  in  the  darkness  as  near 
the  animal  as  we  could  judge  to  be  well  within  range :  we  then 
set  fire  to  it,  and  stood  ready  for  a  shot ;  but  it  was  no  use  ;  the 
brute,  whatever  it  was,  only  retired  further  in,  growling  away 
more  than  ever. 

The  light  of  the  fire  proved  the  cave  to  be  some  hundred 
feet  high,  gradually  narrowing  to  the  top  from  the  bottom, 
which  near  the  entrance  is  about  12  feet  wide,  thus  showing 
it  to  have  been  formed  by  two  gigantic  rocks  or  cliffs  flung 
against  one  another  at  the  time  these  limestone  mountains  were 
thrown  up  from  the  bottom  of  the  sea,  which  in  remote  ages 
doubtless  flowed  over  them.  On  both  sides  of  the  Straits,  «.  e. 
at  Gibraltar  and  Abyla,  these  fissures  or  caves  are  common  in 
the  limestone  ;  but  this  particular  one  fairly  rivals  the  well- 
known  St.  Michael's  Cave  at  Gibraltar,  and  had,  from  the  marks 
of  fire,  been  inhabited  at  some  not  very  distant  period.  The 
floor  in  places  was  quite  a  foot  deep  with  the  guano  of  Rock- 
Doves  (Columla  lima],  numbers  of  which  flew  out  from  the 
nooks  and  crannies  of  the  rock. 

As  far  as  we  could  understand  from  the  Moors,  who,  living 
near  Ceuta,  spoke  a  few  words  of  broken  Spanish,  there  was 
a  story  of  a  Moor  having  ascended  this  cave  till  he  came  out 
somewhere  at  the  top  of  the  mountain  ;  be  this  as  it  may, 


12  INTRODUCTION. 

there  was  a  fine  breezy  draught  of  air  blowing  downwards, 
which  sent  the  smoke  of  our  fire  towards  us  till  we,  instead 
of  the  beast  for  whose  benefit  it  was  intended,  were  nearly 
suffocated. 

Having  no  means  of  getting  torches  to  further  explore  the 
cave,  with  heavy  hearts  we  left  the  unknown  animal  to  growl 
himself  to  sleep ;  the  Moors  insisted,  by  the  way,  that  what 
we  heard  was  a  "  djinn,"  or  evil  spirit! 

The  south-east  part  of  the  range  of  the  Sierra  Bullones  is  a 
different  formation,  and  we  could  trace  signs  of  lead-ore  and 
antimony  in  more  than  one  place.  Whether  any  mines  will  ever 
be  worked  in  Morocco  is  doubtful :  there  is  plenty  of  ground ; 
but  at  present  it  is  forbidden  to  look  for  minerals  by  the 
enlightened  and  despotic  Moorish  government. 

The  track  or  road  from  Ceuta  to  Tetuan,  after  quitting  the 
mountainous  district,  passes  through  the  interminable  scrub 
usual  to  the  Mediterranean  coast ;  and  bad  as  are  mountain 
tracks  in  Spain,  this  one  beats  them  all  in  roughness ;  and, 
owing  to  the  weary  sameness  of  going  up  and  down  hill  after 
hill,  the  journey  seems  endless. 

En  route,  however,  by  the  shore,  nearly  opposite  Tarifa,  lies, 
shrouded  in  large  thick  bushes,  the  ruins  of  Alcazar  Leguer,  a 
large  old  castellated  Portuguese  fort,  built  about  the  beginning 
of  the  fifteenth  century.  Some  parts  of  the  walls  are  in  fair 
condition ;  but  the  interior  is  very  much  dilapidated,  and  the 
whole  overgrown  with  wild  olive-  and  fig-trees,  brambles  and 
rubbish,  desolation  beyond  description,  its  only  tenants  being 
Owls  and  (say  the  Moors)  evil  spirits.  A  covered  way,  formed 
by  two  parallel  high  walls  with  banquettes  on  their  tops,  runs 
down  to  the  sea-shore,  where  it  is  broken  down  and  blocked  up 
with  sand ;  the  ruins  show  signs  of  unskilful  workmanship,  and 
contrast  very  unfavourably  with  those  of  Roman  construction, 
besides  which,  from  being  principally  built  of  soft  sandstone, 
they  are  much  weather-worn  where  exposed  to  the  rain. 


MIGRATION.  13 

Wherever  I  have  wandered  about  in  the  coast-districts  of 
Morocco  the  country  is  singularly  destitute  of  trees  of  any  size, 
what  few  there  are  being  in  the  santos  or  graveyards.  The 
consequence  of  this  is,  there  is  no  change  in  the  landscape ; 
stunted  bushes,  rocks,  and  cultivation  constitute  the  general  view. 
Nevertheless  the  climate  is  splendid  and  healthy,  perhaps  better 
than  that  of  Andalucia ;  and  one  quits  it  with  the  regret  that 
such  a  fine  country  should  in  these  days  of  civilization  be,  as  it 
were,  utterly  wasted — a  land  rich  beyond  most  in  soil,  minerals, 
and  natural  advantages  of  all  sorts,  within  four  days  of  England, 
remaining  without  any  real  government,  without  roads,  bridges, 
or  any  means  of  communication,  owing  to  political  necessities 
abandoned  to  barbarians,  whose  chief  object  seems  to  be  to  keep 
the  country  as  much  as  possible  secluded  from  the  prying  eyes 
of  Europeans. 

MIGRATION,  SHOOTING,  &c. 

The  migration  of  birds,  although  a  most  interesting  subject, 
is  yet  very  imperfectly  understood,  and  reliable  data  from 
different  countries  and  places  are  still  greatly  wanted  to  elucidate 
it.  Without  doubt  caused  by  the  absence  or  abundance  of 
food,  which  in  turn  is  caused  by  difference  of  temperature, 
the  passage  of  birds  in  these  parts  begins  with  most  species 
almost  to  a  day  in  the  spring,  usually  lasting  for  about  three 
weeks,  though  some,  as  the  Hoopoe  and  the  Swallows,  are  more 
irregular  in  their  first  appearance  ;  and  with  these  the  migration 
lasts  throughout  a  longer  period. 

Few,  indeed  hardly  any  birds,  do  not  migrate  or  shift  their 
ground  to  some  extent.  I  can  name  very  few  which  do  not 
appear  to  move,  viz.  Griffon  Vulture,  Imperial  Eagle,  Eagle- 
Owl,  Blue  Thrush,  all  the  Woodpeckers,  Tree-Creeper,  Black- 
headed  Warbler,  Dartford  Warbler,  Crested  Lark,  Chough, 
Haven,  Magpie,  Red-legged  and  Barbary  Partridges,  and  the 


14  INTRODUCTION. 

Andalucian  Quail.  Generally  speaking,  it  seems  to  me  that 
in  the  vernal  migration  the  males  are  the  first  to  arrive,  as 
with  the  Wheatears,  Nightingales,  Night-Herons,  Bee-eaters ; 
but  this  is  a  theory  which  requires  more  confirmation.  Some 
species,  as  the  Neophron  and  most  of  the  Raptores,  pass  in 
pairs. 

Most  of  the  land-birds  pass  by  day,  usually  crossing  the 
Straits  in  the  morning.  The  waders  are,  as  a  rule,  not  seen  on 
passage ;  so  it  may  be  concluded  they  pass  by  night,  although 
occasionally  Peewits,  Golden  Plover,  Terns,  and  Gulls  have  been 
noticed  passing  by  day. 

The  autumnal  or  return  migration  is  less  conspicuous  than 
the  vernal:  and  whether  the  passage  is  performed  by  night, 
or  whether  birds  return  by  some  other  route,  or  whether  they 
pass  straight  on,  not  lingering  by  the  way  as  in  spring,  is  an 
open  question  ;  but  during  the  autumn  months  passed  at 
Gibraltar  I  failed  to  notice  the  passage  as  in  spring,  though 
more  than  once  during  the  month  of  August  myself  and  others 
distinctly  heard  Bee-eaters  passing  south  at  night,  and  so 
conclude  other  birds  may  do  the  same. 

We  have  (vide  Andersson's  '  Birds  of  Damara  Land,'  pp.  18- 
21)  an  account  of  the  swarms  of  Hawks  which  appear  there  at 
the  time  they  are  absent  from  Europe  and  North  Africa ;  so  it 
may  be  reasonably  inferred  with  regard  to  one  species,  Milvus 
migrans  (the  Black  Kite),  that  some  of  the  vast  numbers  which 
pass  the  Straits  of  Gibraltar  retire  in  autumn  through  the  tropics 
to  South  Africa. 

The  best  site  for  watching  the  departure  of  the  vernal 
migration  is  at  Tangier,  where  just  outside  the  town  the  well- 
known  plain  called  the  "  Mashan,"  a  high  piece  of  ground 
that  in  England  would  be  called  a  common,  seems  to  be  the 
starting-point  of  half  the  small  birds  that  visit  Europe. 

Both  the  vernal  and  autumnal  migrations  are  generally 
executed  during  an  easterly  wind,  or  Levanter :  at  one  time  I 


SHOOTIXG.  15 

thought  that  this  was  essential  to  the  passage ;  but  it  appears 
not  to  be  the  case,  as  whether  it  be  an  east  or  west  wind,  if  it 
be  the  time  for  migration,  birds  will  pass,  though  they  linger 
longer  on  the  African  coast  before  starting  should  the  wind  be 
westerly ;  and  all  the  very  large  flights  of  Raptores  (Kites, 
Neophrons,  Honey-Buzzards,  &c.)  which  I  have  seen  passed 
with  a  Levanter.  After  observing  the  passage  for  five  springs 
I-  am  unable  to  come  to  any  decided  opinion,  the  truth  being 
that  as  an  east  wind  is  the  prevalent  one,  the  idea  has  been 
started  that  migration  always  takes  place  during  that  wind. 
Nevertheless  it  is  an  undoubted  fact  that  during  the  autumnal 
or  southern  migration  of  the  Quail  in  September,  they  collect 
in  vast  numbers  on  the  European  side,  if  there  be  a  west  wind, 
and  seem  not  to  be  able  to  pass  until  it  changes  to  the  east ; 
this  is  so  much  the  case  that,  should  the  wind  keep  in  that 
quarter  during  the  migration,  hardly  any  are  to  be  seen. 

On  some  occasions  the  passage  of  the  larger  birds  of  prey  is 
a  most  wonderful  sight ;  but  of  all  the  remarkable  flights  of  any 
single  species,  that  of  the  Common  Crane  has  been  the  most 
noteworthy  that  has  come  under  my  own  observation. 

On  the  Andalucian  side  the  number  of  birds  seen  even  by 
the  ordinary  traveller  appears  strikingly  large,  this  being,  no 
doubt,  in  a  great  measure  caused  by  the  quantity  which  are, 
for  ten  months  at  least  out  of  the  year,  more  or  less  on  migra- 
tion ;  that  is  to  say,  with  the  exception  of  June  and  July,  there 
is  no  month  in  which  the  passage  of  birds  is  not  noticeable, 
June  being  the  only  one  in  which  there  may  be  said  to  be 
absolutely  no  visible  migration,  as  during  the  month  of  July 
Cuckoos  and  some  Bee-eaters  return  to  the  south. 

Though  shooting  is  hardly  a  subject  within  the  design  of 
an  ornithological  brochure  like  the  present,  yet  it  generally 
happens  that  an  ornithologist  is  also  a  sportsman ;  and  there- 
fore a  few  lines  on  the  subject  may  be  acceptable. 

In   the    coast-districts    of  Morocco  no  large  game  is  found 


16  INTBODUCTION. 

within  reach  of  the  European  sportsman,  excepting  wild  pigs, 
which  are  only  to  be  obtained  by  the  battue  system  of  driving 
the  jungle  with  beaters  and  dogs,  sitting  for  hours  waiting  for  the 
chance  of  a  shot,  a  class  of  amusement  dignified  by  the  name  of 
a  "  boar-hunt " ;  sometimes,  where  the  country  is  sufficiently 
open,  the  real  sport  of  pig-sticking  can  be  had. 

No  doubt  further  in  the  interior  there  is  other  large  game  ; 
but  with  the  exception  of  shooting  an  occasional  gazelle  and  a 
few  pigs,  there  is  no  opportunity  of  using  the  rifle. 

The  small-game  shooting  is  very  good ;  the  abundance  of 
Barbary  Partridges  in  some  districts  is  miraculous  ;  but  when 
killed  they  are  of  little  value  in  a  culinary  point  of  view,  being 
more  dry  and  tasteless  than  the  Spanish  Redleg  (Caccabis  rufa], 
now  SQ  well  known  in  many  parts  of  England. 

The  number  of  Snipe  in  some  seasons  is  very  great,  especially 
at  Meshree  el  Haddar,  where,  and  also  at  Ras  el  Doura,  Larache, 
Sharf  el  Akab,  Martine  near  Tetuan,  and  Esmir  near  Ceuta,  as 
good  snipe-  and  wildfowl-shooting  as  may  be  wished  for  can 
be  obtained.  But  it  is,  as  ever  in  Europe  with  Snipe,  always 
uncertain  sport,  as  one  day  swarms  are  met  with,  and  perhaps  on 
the  next  day  hardly  any  are  to  be  found.  The  absence  of  roads 
and  bridges  renders  the  country  in  wet  weather  at  times  impossible 
to  travel  over,  the  tracks  becoming  a  succession  of  mudholes, 
and  the  rivers  impassable  torrents.  This,  added  to  the  unpleasant 
certainty  of  living  under  canvas  during  rainy  weather,  is  a  great 
drawback  to  winter  shooting. 

Another,  in  my  opinion  insuperable,  objection  to  shooting  in 
Morocco  is,  that  if  any  great  quantity  of  game  be  bagged,  it 
has  to  be  thrown  away,  as,  unless  within  twenty  miles  or  so  of 
Tangier,  it  is  useless.  The  Moors,  being  Mahometans,  will 
not  eat  any  thing  killed  by  a  Christian  or  infidel ;  and  killing 
for  the  mere  sake  of  slaughter  does  not  come  within  the  creed 
of  a  true  sportsman.  In  Spain  all  this  is  very  different,  as  any 
one  and  every  one  is  only  too  glad  to  accept  of  the  surplus  game. 


GAME-SHOOTING.  17 

In  many  parts  of  Morocco  rabbits  abound ;  and  hares  are  in 
places  plentiful.  Woodcocks  are  sometimes  tolerably  abundant ; 
Quails,  of  course,  are  in  swarms  during  migration ;  and  there 
are  a  great  number  of  Little  Bustard. 

Shooting  in  Andalucia  is  far  more  satisfactory  and  pleasant 
sport  than  on  the  African  side.  In  the  first  place,  accommodation 
can  always  be  had  in  a  house  of  some  sort,  which  in  warm 
weather,  however,  usually  swarms  with  fleas  ;  but  by  taking  your 
own  blankets  and  a  camping-palliasse,  which  can  be  refilled  at 
each  resting-place  with  chopped  straw,  one  can  generally,  by  the 
aid  of  a  liberal  use  of  either  flea-powder  *  or  albo-carbon 
(naphthalin),  manage  to  cheat  the  vermin  of  their  nocturnal 
banquet.  It  is  almost  absolutely  necessary  to  take  this  powder 
with  one,  as  sleep  in  some  of  the  dens  where  I  have  passed  the 
night  would  have  been  impossible  without  using  it.  Another 
most  useful  item  is  an  india-rubber  flexible  bath,  as  it  is  not 
always  that  a  "  lebrillo  "  or  large  earthenware  pan  big  enough  to 
wash  in  can  be  obtained. 

In  addition  to  the  shelter  to  be  got  in  Andalucia  there  are 
roads ;  and  bad  as  some  may  be,  they  do  afford  means  of 
communication ;  and  there  are  bridges,  though  not  always 
placed  in  the  right  situation;  for  in  places  you  see  a  bridge 
built  across  a  gully  without  any  road  on  either  side  of  it,  and 
others  where  the  stream  has  quitted  its  old  course  for  a  new 
one— single  instances  out  of  the  many  thousand  strange  and 
wondrous  cosas  de  Espana. 

The  large  game  is  more  varied  and  plentiful  in  Andalucia 
than  in  Morocco.  In  most  of  the  wooded  valleys  of  the  sierras, 
near  Gibraltar,  there  were  a  good  many  roe-deer  (corzo)  and  a 
few  wild  pigs ;  in  some  of  the  high  sierras  near  Ronda,  Ubrique, 
and  in  the  Sierra  Nevada  the  Spanish  Ibex  is  sparingly  found*; 

*  This  vegetable  powder  .is  made  from  a  species  of  Feverfew  (Pyrethrum 
roseum),  and  is  quite  innocuous  except  to  insects  ;  many  other  plants  of  the 
Chrysanthemum  group  are  equally  offensive  to  parasitic  insects, 

C 


18  INTRODUCTION. 

but  it  is  extremely  difficult  to  get  them  without  organizing  a 
regular  drive  or  batida — a  very  expensive  affair,  requiring  a 
party  of  several  guns,  who  must  take  tents,  cooks,  &c.  up  into 
the  mountains ;  and  then,  if  successful,  as  far  as  sport  is 
concerned  it  is  hardly  worth  while  sitting  for  several  hours 
behind  a  stone,  nine  times  out  of  ten  without  even  seeing  an 
ibex.  It  is  very  difficult  to  stalk  them,  as  they  lie  hidden  in 
the  thick  stunted  fir  and  other  scrub  which  is  scattered  in  large 
patches  on  the  mountain-sides,  and  are  so  wary  that  you  cannot 
come  suddenly  on  them  like  roe-deer.  However,  in  an  ibex- 
shooting  expedition,  one  is  amply  repaid  by  the  magnificent 
scenery  and  the  novelty  of  the  affair  ;  but  as  far  as  shooting  goes 
it  is  a  failure,  and  every  ibex  killed  by  a  Gibraltar  party  costs 
more  than  I  should  like  to  state. 

Ibex    drop    their   young   about   the    end    of  April ;    on   one 

occasion    a    shooting    expedition    with    which    I    was    present 

succeeded  in  getting  two,  both   of  which  I  sent  home  to  the 

Zoological    Gardens ;     but    unfortunately   they    did    not    long 

survive. 

I  am  informed  that  "  Ibex,  as  late  as  1830,  frequented  the 
sierras  above  Algeciraz,  but  a  disease  which  broke  out  among 
the  tame  goats  was  communicated  to  the  Ibex,  who  all  perished." 

In  the  Sierra  Morena,  near  Palma,  a  little  to  the  west  of 
Cordova,  are  red  deer  strictly  preserved  and  well  pastured  ;  the 
"  heads  "  of  the  stags  are  very  fine,  which  is  not  the  case  with 
those  of  the  Goto  Dofiana,  near  San  Lucar  de  Barrameda. 
All  these,  however,  being  wood-frequenting  deer,  the  antlers 
do  not  branch  out  very  widely,  most  of  the  heads  being  rather 
narrow.  It  is  in  small-game  shooting  that  Andalucia  excels, 
though  it  is  in  no  way  equal  to  that  of  the  countries  lying 
east  of  the  Mediterranean.  Foremost,  both  in  numbers  and 
sport,  is  Snipe-shooting ;  for  in  some  seasons,  about  November 
and  December,  if  the  weather  has  been  dry,  it  is  equal  to  any 
that  can  be  obtained  ;  but  all  depends  upon  the  weather,  which, 


SNIPE-SHOOTING.  19 

if  wet,  causes  the  birds  to  disperse  over  the  whole  country,  while 
if  it  be  dry  they  remain  in  the  sotos  or  marshes,  and  when 
flushed  return  almost  immediately.  Some  of  the  best  sport  I 
have  had  with  them  was  by  waiting  in  favourite  ground  while 
they  kept  coming  in,  flying  high  up  overhead,  and  then  swooping 
down  and  pitching  within  a  few  yards.  Fifty  couple  have  been 
bagged  in  a  day  by  one  gun,  and  that  a  muzzle-loader,  thirty  or 
twenty-eight  couple  a  gun  per  day  being  often  obtained.  The 
proportion  of  Jack-Snipe  is  about  the  same  as  in  England, 
and  they  keep  to  the  most  wet  and  muddy  spots.  Snipe, 
as  a  rule,  in  Andalucia  are  far  wilder  than  in  other  countries, 
which  is  no  doubt  caused  by  the  nature  of  the  marshes, 
which,  often  quite  dry  at  the  end  of  summer,  are  in  winter 
regular  lakes,  only  at  their  edges  affording  any  resting-places 
for  the  birds,  the  cover  being  usually  thin  and  bare. 

There  are  many  acres  of  ground  flooded  with  water,  from 
about  six  inches  to  a  foot  in  depth,  the  whole  dotted  over 
with  tussocks  standing  an  inch  or  two  above  the  water,  and 
about  a  foot  apart  from  each  other.  This  tussocky  ground 
is  most  difficult  both  to  walk  over  and  shoot  on,  as  the  tufts 
are  not  broad  enough  to  stand  on  with  both  feet,  and  these 
slippery  lumps  of  mud  and  grass  standing  above  the  water 
enable  the  Snipe  to  see  a  long  distance,  and  cause  them  to 
rise  very  wildly  ;  while  they  also  have  a  most  provoking  habit 
of  flying  up  just  as  you  are  trying  to  balance  yourself  on  one 
of  the  tussocks.  The  result,  if  you  fire,  is  most  probably  a  miss, 
and  down  you  slip  into  the  water,  lucky  if  on  your  legs  and 
not  on  your  knees  or,  as  happened  to  me  more  than  once,  on 
your  face.  There  is,  however,  one  point  in  favour  of  all  these 
sotos  :  they  have  a  firm  bottom,  the  mud  is  never  deep,  and 
there  are  no  quaking  bogs  or  dangerous  morasses  as  in  Ireland. 
A  retriever,  it  is  almost  needless  to  add,  is  perfectly  indispensable 
for  this  kind  of  sport,  saving  (in  addition  to  many  birds  that 
would  otherwise  be  lost)  much  time  and  the  bad  temper  which 

c2 


20  INTRODUCTION. 

results  from  not  being  able  to  find  birds  that  have  fallen.  Snipe 
in  Andalucia  are  very  seldom  seen  together  in  lots  or  wisps, 
though  occasionally  in  very  wet  stormy  weather  small  wisps 
appear.  The  best  localities  which  I  have  visited  in  Andalucia 
are  the  marshes  near  the  edges  of  the  Marisma,  or  delta  of  the 
Guadalquivir,  below  Seville,  especially  just  beyond  Coria  del 
Rio,  and  near  the  Goto  del  Rey  and  the  Goto  Dofiana ;  one  spot 
near  the  Palacio  of  the  former  place,  las  Carnicerias,  is  excellent. 
At  Casa  Vieja,  or,  more  properly  speaking,  Casas  Viejas,  some 
forty  miles  from  Gibraltar,  is  very  good  ground,  particularly  in 
the  first  part  of  the  season ;  there  are  also  good  marshes  near 
Vejer.  Late  in  the  season,  near  Taivilla  and  Tapatanilla,  on  the 
road  from  Tarifa  to  Vejer,  at  times  Snipe  are  also  to  be  found 
very  plentifully,  but  are  very  wild,  and  it  is  impossible  to  make 
a  large  bag  as  there  is  no  cover. 

The  wildfowl-  or  duck-shooting  in  dry  seasons  is  very  fair 
in  the  early  part  of  the  winter,  before  the  lagoons  and  rivers 
are  filled  up  by  the  rains,  there  being  then  very  few  wet  spots, 
and  the  birds  crowd  together  in  the  small  pools  Avhich  remain 
between  the  high  banks  of  the  river-beds,  and  can  be  easily 
approached ;  but  later  on,  when  these  streams  are  brimful  or, 
rather,  overflow  their  banks,  and  when  the  lagunas  are  sheets 
of  water  without  rushes  or  cover  of  any  sort  at  the  edges,  it 
is  almost  impossible  to  shoot  ducks  by  day  except  by  making 
"  hides  "  with  sticks  and  stones,  and  sending  some  one  round 
and  trying  to  have  them  driven  over  you.  At  flight  sometimes 
very  fair  sport  is  to  be  had  for  one  or  two  nights ;  but  after  that 
the  fowl  know  the  place,  and  either  come  very  late  or  avoid  it 
altogether.  For  flight-shooting  a  good  retriever  is  absolutely 
necessary ;  for  it  is,  in  the  dark,  impossible  to  find  the  spoil ; 
and  if  left  till  morning,  the  Marsh-Harriers  are  at  them  by  break 
of  day,  leaving  nothing  but  bones  and  feathers.  To  my  mind 
there  is  very  great  charm  in  flight-shooting,  and  a  naturalist 
while  waiting  will  see  and  hear  much  that  is  pleasant  and  perhaps 


WILD  GEESE.  21 

some  that  is  new  to  him.  This  sport  requires,  too, .considerable 
skill  in  judging  the  distance,  and  sharpness  of  vision  in  being 
able  to  catch  a  glimpse  of  the  ducks  as  they  pass  over.  It  is  a 
great  help  if  you  can  place  yourself  so  that  you  face  the  west, 
and  thus  get  the  birds  in  the  evening  light,  when  they  can 
be  seen  coming  a  very  long  way  off";  but  if  they  come  from 
the  eastward,  and  you  are  obliged  to  face  that  way,  they  never 
show  till  close  on  you ;  and  the  croaking  frogs  make  an  almost 
deafening  noise,  so  that  you  cannot,  as  in  England,  hear  the 
sound  of  the  ducks'  wings. 

Immense  numbers  of  Wild  Geese  in  some  years  are  found  in 
the  winter  months  about  the  Laguna  de  la  Janda,  and  below 
Seville,  in  the  marshes  of  the  Guadalquivir.  They  are  of 
course  very  difficult  to  "  get  at";  but  as  they  pass  the  day 
on  the  ground  at  the  edge  of  the  water,  and  always  have 
certain  favourite  spots  to  which  they  resort,  they  are  to  be 
got  by  digging  or  making  "  hides "  at  the  places  they  most 
frequent.  In  the  morning,  at  sunrise,  they  collect  on  the 
water,  in  some  places  in  hundreds,  and  swim  about  feeding 
for  an  hour  or  two  on  some  substance  which  they  pick  up 
from  the  bottom  of  the  shallow  water ;  after  this  they  disperse 
and  take  to  the  shore,  where,  if  left  undisturbed,  they  pr.cs 
the  day  sleeping  and  pluming  themselves.  There  is  one  of 
these  goose-haunts  near  the  Palacio  of  the  Goto  del  Rey,  a 
little  to  the  south-east  of  it.  One  morning  in  January,  having 
the  day  previously  made  a  hide  among  some  tufts  of  rushes, 
I  went  and  laid  up  before  sunrise  to  await  the  geese,  which 
arrived  by  degrees  in  flight  after  flight,  till  there  must  have 
been  within  a  mile  of  me,  at  the  lowest  computation,  between 
three  and  four  thousand  ;  I  shall  never  forget  the  sight,  and 
I  lay  concealed  watching  them  for  at  least  two  hours.  I  could 
not  distinguish  amongst  them  more  -than  one  lot  of  about  a 
dozen  Bean-Geese ;  the  remainder  were  all  Grey-lags.  Some 
hundreds  were  within  about  a  hundred  yards,  and  it  was 


22  INTRODUCTION. 

very  amusing  to  see  them  feeding,  fighting,  and  playing  with 
one  another  ;  but  somehow  they  were  evidently  suspicious 
of  the  patch  of  spiky  rushes  in  which  I  was  lying  flat  in  the 
slight  hole  which  had  been  made  between  two  tufts  of  rushes 
and  covered  over  with  others  dug  up  by  the  roots,  and  arranged 
so  as  to  look  as  if  growing.  Unable  to  turn  on  my  side  or 
move  in  the  least,  I  was  so  cramped  that  it  was  all  I  could 
do  to  remain  there ;  but  after  a  time  a  large  lot  of  geese  began 
to  set  in  towards  my  position,  and  in  a  few  minutes  more  I 
should  have  had  a  good  family  shot.  I  had  plenty  of  chances 
of  firing,  but  could  not  have  got  more  than  a  couple ;  besides 
which  I  liked  to  watch  them,  so  waited  on  in  hopes  of  a 
good  lot  coming  close  to  me,  when,  alas !  cries  of  alarm  were 
heard  from  the  birds  furthest  away  on  my  right,  and  after  a 
minute  or  two  they  began  to  fly  up,  and  I  could  see  against 
the  sky  a  man  riding  towards  them.  The  geese  in  front  of 
me  all  pricked  up  their  heads  and  were  getting  ready  to  be 
off;  so  I  was  obliged  to  jump  and  send  both  barrels  at  them 
as  my  only  chance ;  and  by  good  luck,  or  rather  thanks  to 
the  large  shot,  two  were  killed,  but  not  enough  to  recompense 
one  for  lying  cramped  up  for  so  long  ;  nevertheless  I  was  more 
than  repaid  by  the  sight  of  so  many  wild  geese  and  some  other 
fowl  close  to  me,  and  being  able  to  watch  their  movements. 
Any  one  who  would  take  the  trouble  to  try  punt-shooting  with  a 
big  gun  below  Seville  on  tidal  waters  might  make  some  wonderful 
bags,  as  the  enormous  quantities  of  Geese,  Wigeon,  and  other 
ducks  can  only  be  approached  with  the  aid  of  a  punt.  When 
near  the  edge  of  water  you  can  always  approach  Ducks  with  a 
stalking-horse ;  and  Geese  will  allow  this  on  their  first  arrival, 
but  soon  become  too  wary. 

Golden  Plover  are  extremely  abundant  in  vast  flocks  from 
November  to  March.  On  their  first  arrival  they  are  not  so 
wild  as  afterwards.  They  can  always  be  "  got  at "  with  a 
stalking-horse ;  but  as  good  a  plan  to  shoot  them  as  any  is 


QUAILS  AND  BUSTARDS.  28 

to  stand  still  in  some  place  which  they  frequent  .on  a  windy 
day,  when  they  will  often  fly  within  a  few  yards.  Peewits 
are  numerous,  but  not  worth  shooting,  as  is  the  case  with 
Curlews ;  but  the  latter  are,  as  elsewhere,  much  too  wary 
to  allow  themselves  to  be  shot,  and  during  the  whole  time 
I  was  in  Andalucia  I  never  but  once  had  the  chance  of 
killing  one. 

Woodcocks  in  some  seasons  are  numerous  ;  but  five  or  six 
couple  in  a  day  is  a  very  good  bag,  very  different  from  Albanian 
shooting.  Red-legged  Partridges  (Caccabis  rufd)  are  not  worth 
the  trouble  of  going  after,  either  for  sport  or  for  the  table  ;  in  some 
places  there  are  a  good  number,  but  not  near  Gibraltar ;  they 
are  the  chief  object  of  sport  with  the  Spaniards,  who  kill  them 
at  all  seasons,  even  shooting  them  from  the  nest.  Quails  are, 
during  the  enfrada  or  autumnal  migration,  so  extremely  abundant 
that,  if  there  has  been  a  westerly  wind  for  a  few  days  in 
September,  when  they  are  on  passage,  there  is  really  no  limit  to 
the  number  that  may  be  shot.  About  Tarifa  at  that  season 
every  gun-possessing  man  and  boy  turns  out  with  all  the  cur 
dogs  in  the  town,  and,  regardless  of  each  other,  they  fire  in 
all  directions,  so  that  it  is  a  service  of  danger  to  go  out  near 
them.  If  the  wind  during  their  passage  remains  in  the  east, 
the  Quails  pass  on,  and  little  or  no  sport  is  to  be  had  with 
them.  A  west  wind  appears  to  detain  them  and  prevent  their 
passing  the  Straits,  though  it  does  not  seem  to  retard  their 
migration  by  land. 

The  remaining  small  game  to  be  noticed  in  Andalucia  are 
Bustards,  hares,  and  rabbits.  The  Great  Bustard  is  only  to 
be  got  with  any  certainty  by  driving.  The  Little  Bustard, 
more  wary  still,  is  only  to  be  shot  in  the'  end  of  July  and 
in  August  during  the  extreme  heat  of  the  day,  though  rarely 
they  can  be  driven  over  a  gun  by  getting  under  the  bank 
of  a  river  or  such  like  shelter,  and  sending  a  man  round 
to  put  them  up  ;  but  on  rising  they  usually  mount  up  very 


24  INTRODUCTION. 

high,  in  this  respect  differing  from  the  Great  Bustard,  which 
seldom  flies  high  enough  to  be  out  of  shot  if  you  are  directly 
underneath.  Hares  (Lepus  mediterraneus)  are  a  much  smaller 
species  than  in  England,  about  the  size  of  a  good  average 
English  rabbit,  not  very  abundant  anywhere  and  frequenting 
open  fiat  and  cultivated  districts,  never  being  found  among 
woods  or  high  ground.  Rabbits,  of  course,  are  abundant 
but  very  small,  rather  less  in  size  than  the  New  -  Forest 
rabbit,  which  is  the  most  diminutive  race  in  England.  A 
shooting  -  license,  easily  obtainable  through  the  aid  of  any 
British  Consul,  is  requisite  in  Spain ;  and  though  seldom 
asked  for,  it  is  better  to  have  one.  The  form  and  cost  of 
one  varies  according  to  the  Government,  and  therefore  is 
seldom  alike  two  years  in  succession.  A  close  time  has 
been  established,  but,  needless  to  say,  in  the  wild  districts  is 
not  regarded. 

Here  follows  a  partial  list  of  the  Mammalia  of  Andalucia, 
with  their  local  names,  which  may  be  useful  to  the  sportsman. 
Of  course  there  are  other  species  of  small  Mammals  to  be  found, 
especially  among  the  Bats ;  with  the  names  of  the  latter  I 
have  been  kindly  assisted  by  Lord  Lilford,  who  has  personally 
obtained  them  all  in  Andalucia.  Those  marked  with  an  asterisk 
I  have  obtained  myself  or  seen  "  in  the  flesh." 

Greater  Horse-shoe  Bat  Ehinolophus  ferrum-equinum. 

*  E.  euryale. 

Lesser  Horse-shoe  Bat .  E.  bibastatus. 

Dysopes  rueppellii. 

Barbastelle Barbastellus  communis. 

*Noctule    Vespertilio  noctula. 

*Mouse-coloured  Bat   . .  V.  tnurinus. 

*Schreiber's  Bat    V.  schreiberi. 

Long- eared  Bat V.  auritus. 

V.  schreiberi,  V.  murinus,  and  R.  euryale  are  found  in  caves 
near  Casas  Viejas — the  two  former  species  in  countless  numbers, 


MAMMALIA. 


25 


the  dung  at  the  bottom  of  the  caves  being  from  four  to  five  feet 
in  depth.     The  Spanish  name  for  all  is  Murcielayo. 


*Hedgehog Erinaceus  europseus. 

*Shrevv Sorex  araneus. 

*Mole    Talpa  europaea. 

*Badger     Meles  taxus. 

*Comrnou  Marten  Cat . .  Mustela  foina. 

*Polecat     M.  putorius. 

*Weasel     M.  vulgaris. 

*0tter   Lutra  vulgaris. 

*Genet Viverra  genet ta. 

*Ichneumon Herpestes  ichneumon. 

*Wild  Cat      Felis  catus. 

*Spanish  or  Spotted  Lynx  F.  pardina. 

*Wolf    Canis  lupus. 

*Fox C.  vulpes,  var.  melanogaster. 

Squirrel Sciurus  vulgaris. 

*Fat  Dormouse     Myoxus  glis. 

M.  nitella. 

*Dormouse    M.  avellanarius. 

*Brown  Eat Mus  decumanus. 

*Mouse      M.  musculus. 

*Black  Eat    M.  rattus. 

*Long-tailed  Field-Mouse  M.  sylvaticus. 

*Water-Eat Arvicola  amphibius. 

*Field-Mouse    A.  agrestis. 

*Hare    Lepus  mediterraneus. 

*Eabbit      L.  cuniculus. 

*Wild  Pig     Sus  scrofa. 

*Eed  Deer     Cervus  elaphus. 

*Fallow  Deer    C.  daina. 

*Eoe  Deer     C.  capreolus. 

*Ibex     Capra  hispanica. 


Spanish  names. 
Erizo. 
Musarana. 
Topo. 
Tejon. 
Foina. 
Turon. 
Coinadreja. 
Nutra  or  Nutria. 
Gineta. 

Melon,  Meloncillo. 
Gato  raontes. 
Gato  clavo,  Gato  cerval. 
Lobo. 
Zorro. 
Ardilla. 

Liron  campestre. 
Eaton  careto. 
Lirori  de  los  Avellanos. 
Eata. 
Eaton. 
Eata  negro. 
Eaton  de  campo. 
Eata  de  agua. 
Topino. 
Liebre. 
Conejo. 

Jabali,  Jabalina. 
Ciervo. 

Gamo,  Paleto. 
Corzo. 
Cabra  monte's. 


In  this  book  I  have  endeavoured  to  name  with  each  species 
of  bird  some  definite  locality  where  they  may  be  found,  which 
is  rather  necessary,  as  certainly  on  the  Spanish  side  of  the 
Straits  birds  are  very  locally  distributed,  perhaps  more  so  than 


26  INTRODUCTION. 

in  most  countries  I  have  visited.  It  is  difficult  to  surmise  the 
cause  of  this, .as  precisely  similar  tracts  of  country  within  no 
very  great  distance  of  each  other  are  not  always  frequented 
by  the  same  birds.  On  the  Spanish  side,  without  doubt,  the 
most  common  bird  as  regards  numbers  is  the  Goldfinch,  and 
the  most  universally  distributed  the  Stonechat.  The  number 
of  birds  of  prey  is  very  great,  chiefly  feeding  on  rabbits,  rats, 
mice,  reptiles,  and  insects ;  they  are  very  useful,  and  as  the 
ground-breeding  birds  suffer  much  in  the  nesting-season  from 
snakes  and  lizards,  those  birds  of  prey  which  feed  chiefly  on 
these  enemies  of  the  smaller  birds  render  their  lesser  brethren 
valuable  protection.  The  number  of  little  birds,  especially 
during  the  season  of  migration,  is  sure  to  be  noticed  even  by 
the  most  unobservant.  Immense  quantities  of  Larks,  Finches, 
and  even  some  of  the  Warblers  are  brought  into  the  markets ; 
but  as  a  Spaniard  seldom  shoots  at  such  small  fry,  they  are 
chiefly  netted,  caught  at  night  with  a  lantern  and  bell,  or  snared 
with  bird-lime  (%#). 

The  best  localities  for  an  ornithologist  living  at  Gibraltar 
to  obtain  specimens  or  watch  migration  is  the  country  west 
of  an  imaginary  line  drawn  due  north  from  Gibraltar  as  far  as 
the  latitude  of  Seville.  Within  this  district,  part  of  which  is 
given  in  the  Map  attached  to  this  volume,  as  much  can  be  done 
as  is  possible  in  three  or  four  months'  time  ;  and  the  district  is 
large  enough  to  require  many  years  to  work  it  out  thoroughly. 

In  the  immediate  vicinity  of  Gibraltar  (or  el  Penon,  as  the 
Spaniards  call  it),  the  Cork-wood  of  Almoraima  and  the  level 
ground,  mud-flats,  and  old  salinas  "  between  the  rivers "  on 
the  way  to  Algeciraz  offer  to  the  collector  capital  ground  for 
work.  In  the  Cork-wood  particularly  several  birds  are  found 
breeding  which  do  not  seem  to  nest  elsewhere.  The  ground 
north-east  of  Gibraltar  is  to  a  great  extent  covered  with  scrub 
and  brushwood ;  and  little  is  to  be  done  in  the  bird  line  in  that 
direction. 


LOCALITIES  FOR  COLLECTING.  27 

The  sierras  being  too  far  distant,  cannot  be  worked  from 
Gibraltar ;  it  is  necessary  to  go  to  Algeciraz,  Facinas,  Pulverilla, 
or  some  cortijo  near  the  hills  you  wish  to  work.  Very  deceptive 
in  appearance,  looking  quite  low  and  easy  to  ascend,  it  takes 
three  or  four  hours  to  reach  their  tops,  which,  bare,  rugged, 
and  wild  beyond  description,  are  alone  worth  visiting  for  the 
view,  which,  always  grand,  on  a  clear  day  is  magnificent,  that 
from  the  Penon  del  Fraile  to  the  west  of  Algeciraz  being  one  of 
the  finest.  From  these  mountains  run  down  numerous  wooded 
valleys  (gargantas)  clothed  with  cork  and  oak  trees,  many  of  very 
large  size,  though  badly  mutilated  by  being  lopped  by  charcoal- 
burners.  The  rocky  streams  which  flow  down  these  valleys  are 
fringed  with  rhododendron,  arbutus,  holly,  hawthorn,  laurestinus, 
oleander,  bay,  myrtle,  giant  heather,  cistus,  and  many  sorts  of 
ferns,  conspicuous  amongst  them  being  the  Osmunda  and  maiden- 
hair, while  here  and  there  is  an  occasional  Caladium  with  its  huge 
leaves  reminding  one  in  shape  of  elephant's  ears  :  the  leaves  of 
this  plant,  called  Jwjas  de  llama,  are  much  used  by  the  country 
people  as  a  medicine  for  fevers  ;  many  of  the  rocks  and  all  the 
trunks  of  the  cork-trees  are  festooned  with  hare's-foot  fern 
(calaguala),  also  used  medicinally. 

In  spring  these  ravines  are,  from  their  natural  beauty  and 
the  colour  of  these  various  shrubs  and  flowers,  so  picturesque 
that  one  cannot  help  lingering  about  them  merely  to  admire  the 
charming  scenery,  becoming  apt  to  forget  the  birds  for  which 
one  is  in  search.  These  places  are  seldom  visited  by  an 
Englishman,  only  by  stray  smugglers,  goatherds,  and  charcoal- 
burners  ;  and  every  pass,  hill,  valley,  in  fact  every  well-marked 
situation,  has  its  name,  many  more  familiar  to  me  than  the 
streets  of  London. 

Those  valleys  most  worth  visiting  near  Gibraltar  are  the 
Garganta  del  Capitan,  to  the  north-west  of  Algeciraz,  on  the 
way  to  Ojen  by  the  mountain-path  of  la  Trocha,  which  is 
within  easy  distance  (five  or  six  miles)  of  Algeciraz.  The  valley 


28 


INTRODUCTION. 


of  the  Guadalmalcil,  halfway  on  the  road  between  Tarifa  and 
Algeciraz,  is  also  very  beautiful ;  but  the  Garganta  del  Helecho 
(Valley  of  the  Ferns),  south-west  of  Pulverilla,  is  perhaps  the 
best  for  shrubs,  flowers,  and  ferns.  The  "  Waterfall "  valley, 
near  Algeciraz  (la  Garganta  del  Aguila),  is  tamer  than  any  ;  but 
above  the  cascades  or  waterfalls  it  improves  on  acquaintance. 
This  ravine,  however,  is  well  knowTn  to  every  one  who  has  been 
at  Gibraltar  as  the  regular  rendezvous  for  picnics,  the  very 
name  of  which  is  enough  to  destroy  any  merits  that  the  scenery 
may  possess.  Towards  Tarifa  and  beyond,  on  the  road  to  Vejer, 
the  country  is  not  so  pretty,  opening  out  near  Facinas  to  the  vega 


TEBEACE  ON  THE  NORTH  SIDE  OF  SAN  BAETOLOM^. 

of  the  Laguna  de  la  Janda ;  thence  cultivated  ground,  or 
campina,  stretches  away  to  Medina  Sidonia  and  on  to  Jerez. 
On  the  right  and  left  of  this  road,  however,  are  three  isolated 
rocky  ranges — those  of  la  Sierra  de  San  Bartolome  and  la  Sierra 
de  la  Plata  being  to  the  left,  that  of  la  Sierra  de  Enmedio  to  the 


EOCKY  CLIFFS,  ETC.  29 

right ;  these  ranges  are  the  breeding-places  of  Griffon  Vultures 
and  other  rock-breeding  birds,  and  are  well  worthy  of  a  visit. 

I  here  give  the  names  of  a  few  of  the  rocky  cliffs  which  should 
be  visited  by  those  who  wish  to  see  such  places  : — la  Laja  de  la 
Zarga  and  la  Silla  del  Papa,  in  the  Sierra  de  Plata ;  la  Laja  del 
Ciscar  *,  to  the  east  of  and  near  Taivilla ;  Piedra  de  Paz,  near 
Paterna ;  la  Laja  de  los  Pajaros,  los  Jolluelos,  and  la  Laja  de 
Peiiarroyo,  near  Casas  Viejas. 

There  are  also  magnificent  cliffs  in  the  Sierra  de  las  Cabras, 
east  of  Alcala  de  los  Gazules,  and  hundreds  of  others  which  I 
saw  but  could  not  find  time  to  visit.  I  did  not  care  to  send 
"collectors"  to  bring  eggs  without  the  birds  to  which  they 
belonged ;  or,  as  is  often  the  case  with  these  worthies,  they 
would  have  brought  eggs  with  birds  to  which  they  did  not  belong, 
and,  with  unblushing  effrontery,  sworn  perhaps,  as  I  have  known 
them  do,  that  a  Turkey's  egg  was  taken  by  them  in  a  high  cliff, 
and  belonged  to  an  "  Aguila  de  las  rocas." 

It  is  to  be  hoped  that  this  book  may  not  be  the  cause  of  the 
useless  or  unnecessary  destruction  of  any  bird,  and  especially 
that  dealers  may  not  profit  thereby.  All  mentioned  is  intended 
for  the  benefit  of  true  ornithologists,  and  not  for  those  who  are 
never  satisfied  unless  killing  or  having  killed  as  many  rare  birds 
as  possible. 

It  will  be  seen  that  there  is  sport  to  be  had  in  Andalucia ; 
and  the  shooting  has  the  charm  of  a  varied  bag,  and  the  freedom 
to  wander  where  you  like,  as  a  rule ;  added  to  which  it  is 
necessary  to  work  for  your  game,  which,  in  my  idea,  adds  much 
to  the  pleasure  of  sport.  The  climate,  too,  is  all  that  can  be 
wished,  especially  in  spring,  when  there  is  something  most 
exhilarating  in  the  air;  but  in  autumn,  until  October,  it  is  too 
warm  to  go  out  with  pleasure,  and  the  sun-baked,  tawny,  dusty, 
thirsty-looking  country  has  lost  all  the  beauty  of  its  flowers  and 
the  verdure  of  spring.  To  see  Andalucia,  it  should  be  visited  in 

*  This  name  is  from  "  ciscar,"  to  besmear,  the  whole  face  of  the  laja  being  so 
whitened  by  th.3  Vultures'  droppings  as  to  be  conspicuous  some  miles  off. 


30  INTRODUCTION. 

March,  April,  and  May,  in  order  to  thoroughly  appreciate  both 
the  climate  and  the  scenery. 

Another  hint  which  I  would  fain  give  is  to  be  as  civil  as 
possible,  and  conform  to  the  customs  of  the  country.  The 
Andalucian  peasant,  courteous  and  polite,  is  at  heart  a  caballero, 
and  very  different  from  the  inhabitants  of  the  towns  ;  at  the 
same  time  he  is  proud  and  independent,  and,  to  humour  him, 
he  must  be  treated  on  terms  of  equality.  Above  all  things 
remember  that  it  is  no  use  attempting  to  hurry  in  Spain,  where 
patience  is  more  severely  taxed  than  in  any  other  country,  and 
where  no  corre  prieaa  is  the  order  of  the  day.  Certainly  the  best 
cure  for  impatience  is  to  pass  a  few  months  among  Spaniards. 

Here  ends  this  Introductory  Chapter,  with  apologies  for  its 
shortcomings  in  the  fact  that  it  is  the  concoction  of  one  who 
detests  pen,  ink,  and  paper,  and  who  is  more  at  home  with  the 
gun,  rifle,  or  fishing-rod ;  so,  in  the  manner  of  the  country 
which  to  me  has  so  many  charms,  let  me  conclude  with  the 
farewell  and  time-honoured  salutation,  Vaya  Vd.  con  Dws. 


EL  OBGATsO,  KEAB,  TAKIFA. 


CHANGES  IN  THE  VICINITY  OF  THE  ROCK.  31 

Since  the  preceding  was  written,  alas !  more  than  twenty  years 
ago,  the  writer  has  on  several  occasions  in  winter  and  spring 
visited  his  old  haunts,  the  last  expedition  being  made  in  com- 
pany with  Major  Willoughby  Verner,  Rifle  Brigade,  who  was 
quartered  at  Gibraltar  from  1874  to  1881,  and  who  has  since 
made  frequent  visits.  To  him  I  owe  many  records  of  his 
practical  observations,  as  well  as  to  Mr.  Meade-Waldo  for  some 
notes  made  by  him  during  a  visit  to  Tangier  in  the  spring  of 
1892. 

Captain  Savile  Reid,  late  Royal  Engineers,  contributed  a 
paper  to  '  The  Ibis,'  1885,  "  Winter  Notes  from  Morocco " ; 
while  Mr.  Abel  Chapman  wrote  in  '  The  Ibis,'  1884,  "  Rough 
Notes  on  Spanish  Ornithology,"  and  in  many  of  his  Chapters 
of  his  '  Wild  Spain  '  he  treats  of  Andalucian  birds. 

"  Aves  de  Espana/'  by  Don  Jose  Arevalo  y  Baca,  vol.  xi. 
'  Memorias  de  la  Real  Academia  de  Ciencias '  (Madrid,  1887), 
contains  some  information  on  Andalucian  birds,  and  all  given  on 
personal  observation  is  no  doubt  bond  fide,  but  unfortunately  he 
often  quotes  one  upon  whom  we  cannot  rely. 

There  is  now,  in  1894,  little  change  in  the  country  near 
Gibraltar,  except  in  the  dress  of  the  peasantry,  who  have  dis- 
carded most  of  their  picturesque  provincial  costume,  even  the  old 
sombrero  is  rarely  seen,  and  replaced  by  the  felt  hat  of  the 
London  rough  ;  while  the  "  fair  "  sex  have  left  off  the  grace- 
giving  mantilla  and  even  in  wild  out-of-the-way  places  adopted 
caricatures  of  modern  fashions. 

There  seems,  if  possible,  more  poverty,  and  in  the  winter  of 
1893-94  there  was  mucha  hamlre  and  actual  deaths  from  want ; 
and  had  not  the  early  spring  of  1894  been  exceptionally  wet, 
there  would  have  been  a  serious  famine. 

Among  changes  in  the  immediate  vicinity  of  the  Rock,  a 
serious  one  for  the  Garrison  is  that  the  right  of  shooting  over  all 
that  part  "  between  the  rivers  "  to  as  far  as  and  including  the 
Cork-wood  has  been  hired  and  monopolized  by  some  Gibraltar 


32 


INTRODUCTION. 


merchants,  so  the  British  officer,  who  used  to  go  where  he 
wished,  has  no  chance  of  a  day's  gunning  :  he  must  either  cross 
the  Straits  or  make  an  expedition  into  Spain — an  expensive  affair, 
which  takes  a  day  en  route  and  another  on  return. 

It  is  much  to  be  regretted  that  the  Garrison  could  not  have 
arranged  to  hire  this  shooting,  so  as  to  enable  officers  to  have  a 
day's  sport ;  but  the  chance  has  passed  away,  probably  not  to 
return.  In  addition  to  this  there  are  many  other  places  marked 
acotado  (preserved),  which  used  not  to  be  so,  and  many  more 


PEASANTRY, TAPATANILLA. 

guns  are  carried  in  the  country,  which  has  caused  a  great 
decrease  in  the  resident  birds  of  prey  during  the  last  twenty 
years.  About  Gibraltar  the  Griffon  Vulture  is  as  numerous  as 
then,  but  the  Bearded  Vulture  is  gone.  In  the  provinces  of 
Malaga  and  Granada  many  Vultures,  Bearded  and  Griffon,  are 
reported  to  have  been  destroyed  by  poison,  laid  for  wolves  &c. 
As  few  genera  have  been  given  as  possible :  the  present  rage  is 


MEASUREMENTS. 


33 


to  give  as  many  as  any  excuse  can  be  found  for,  and  will  soon 
result  in  every  bird  having  a  separate  genus ;  and  in  addition 
some  writers  now  give  a  bird  the  same  generic  and  specific 
name,  thus  making  as  much  confusion  as  possible  in  nomen- 
clature. 

The  unfortunate  part  of  ornithology,  as  at  present  practised,  is 
that  it  is  chiefly  confined  to  the  slaughter  of  birds,  whose  skins, 
when  compared  and  examined  by  table  naturalists,  are  upon  the 
slightest  variation  in  plumage  made  into  new  species,  without 
any  knowledge  of  their  habits,  notes,  &c.  Much  more  can  be 
done  by  observation  than  by  the  gun,  and  when  a  bird  is 
destroyed  all  chance  of  noticing  its  habits  is  destroyed  likewise. 
Measurements  of  length  given  are  only  an  attempt  to  show  the 
proportionate  size  of  a  species.  Measurements,  in  the  writer's 
opinion,  are  of  little  use,  especially  as  hardly  any  two  persons 
measure  a  bird  in  the  same  manner. 


LAJA  DEL  CISCAB. 


34  TUKD1D.E. 


Order    PASSERES.  Family  TURDIDJE. 

Subfamily  TURBINE.     Young  spotted. 

1.  Turdus  viscivorus,  Linnaeus.    The  Mistle-Thrush. 

Spanish.  Charla  (Chatterer). 

"  Found  near  Tangier,  always  singly  and  very  sparingly  in 
company  with  T.  musicus,  on  passage.  They  arrive  in  November, 
but  do  not  stay  near  here,  returning  to  recross  the  Straits  in 
February.  "—Favier. 

They  occasionally  nest  near  Tangier,  as  in  1869  I  saw  eggs 
taken  near  there. 

The  Mistle-Thrush  cannot  be  said  to  be  common  near 
Gibraltar,  being  most  so  in  winter.  They  are  considered  to 
arrive  and  depart  with  the  Woodcocks  ;  but  a  few  pairs  nest  in 
the  Cork-wood  and  other  wooded  districts. 

Axillaries  white ;  a  white  patch  on  end  of  inner  web  of  two  outer  tail- 
feathers  on  each  side. 

Young.  Spotted  with  buff  and  black  on  head  and  back.  Length 
11  inches. 

2.  Turdus  musicus,  Linnaeus.    The  Song-Thrush. 

Spanish.  Zorzal. 

Favier's  note  applies  to  this  bird  on  both  sides  of  the  Straits, 
and  is  as  follows : — "  The  Song-Thrush  is  a  winter  resident  in 
great  numbers,  being  the  most  common  of  the  Thrushes,  arriving 
in  large  flocks  in  October  and  November,  departing  in  March." 

On  the  Spanish  side  they  chiefly  frequent  the  wild  olive-trees, 
on  the  berries  of  which  they  feed.  The  first  date  of  arrival 


TURDUS  MEEULA.  35 

noticed  at  Gibraltar  was  the  22nd  of  October ;  and  the  latest 
day  on  which  I  observed  them  was  the  1st  of  April. 

Axillaries  biiffish  yellow  ;  eye-stripe  scarcely  developed. 

Young.  Spotted  ;  the  feathers  of  upper  surface  with  pale  centres.     Afte 
first  moult  young  resembles  adult,  but  has  pale  ends  to  the  wing-coverts. 
Length  9  inches. 

3.  Turdus  iliacus,  Linnaeus.    The  Redwing. 

Spanish.  Malvis. 

"This  Thrush  is  very  rare  near  Tangier.  I  have  only  met 
with  two,  between  November  and  March — one  in  1852,  the 
other  in  1864:."— Favier. 

Mr.  Meade-Waldo  observed  them  on  the  17th  of  February, 
1892,  near  Tangier. 

In  Andalucia  the  Redwing  is  abundant  in  winter,  in  company 
with  the  Song-Thrush. 

Axillaries  chestnut-red ;  well-defined  whitish  streak  over  eye,  reaching  to 
nape.  Length  8|  inches. 

4.  Turdus  pilaris,  Linnseus.    The  Fieldfare. 

Is  not  mentioned  by  Favier,  but  Mr.  Meade-Waldo  found 
Fieldfares  in  abundance  on  the  Moorish  side  of  the  Straits 
in  the  spring  of  1892.  On  the  Spanish  side  I  never  met 
with  any,  but  they  are  reported  by  Arevalo  from  Granada. 

Axillaries  white ;  rump  slate-grey.     Length  10  inches. 

5.  Turdus  merula,  Linnaeus.    The  Blackbird. 

Moorish.  Tchau  Tchau  (Favier).     Spanish.  Mirlo. 

"  Resident  near  Tangier  and  very  plentiful,  nesting  three 
times  a  year." — Favier. 

I  found  a  nest  in  Morocco  built  in  a  prickly-pear  hedge. 
The  Blackbird  nests  at  Gibraltar,  and  is  resident  and  very 

D2 


36  TURDID.E. 

common   in   Andalucia :    in  the  winter  months  their   numbers 
are  greatly  augmented  by  migrants. 

Male.  Black  ;  bill  orange-yellow. 

Female.  Brownish  black  or  reddish  brown  ;  bill  brown. 
Young.    Like  female,  but  more  spotted  ;    the  male   with  blackish  bill. 
Length  10  inches. 

6.  Turdus  torquatus,  Linnaeus.    The  Ring-Ouzel. 

Spanish.  Chirlo. 

"  Is  only  met  with  in  small  nights  on  passage  near  Tangier, 
crossing  to  Europe  in  March  and  April,  and  returning  in  the 
autumn  to  pass  the  winter  further  south." — Favier. 

I  only  observed  the  Ring-Ouzel  near  Gibraltar  on  passage 
in  the  spring,  the  earliest  dates  in  each  year  being  the  8th  of 
April  1868,  20th  of  March  1870,  9th  of  April  1871,  12th  of 
March  1872,  28th  of  March  1874;  but  they  are  known  to  breed 
in  the  mountains  near  Granada,  and  Mr.  Saunders  records  a 
nest  near  Colmenar. 

Male.  Uniform  brownish  black,  with  wldte  crescent  on  chest. 
Female.  Lighter  ;  crescent  narrower,  washed  with  brown. 
Young.  Spotted.     Length  11  inches. 

7.  Monticola  cyanus  (Linnaeus).    The  Blue  Rock-Thrush. 

Moorish.  Tchau-tchau  zerak.     Spanish.  Solitario. 

Favier  states  that  the  Blue  Rock-Thrush,  which  is  as  common 
in  suitable  localities  in  Morocco  as  in  Andalucia,  is  migratory, 
passing  north  from  February  to  May,  and  passing  south  from 
August  to  September.  I  never  could  detect  any  migration 
on  the  Spanish  side,  and  consider  it  one  of  the  very  few  birds 
which  are  stationary,  not  even  shifting  their  ground — though, 
perhaps,  in  other  countries  circumstances  may  cause  them  to 
migrate. 

Abundantly  distributed  on  all  rocky  ground,  even  on  sea-cliffs, 
and  often  seen  on  house-tops  in  those  towns  which  lie  in  their 
districts,  they  are  always  to  be  found  at  Gibraltar  in  unvarying 


MOxNTICOLA  CYAN  US.  37 

numbers,  frequenting  daily  the  same  spots,  and  attracting 
considerable  notice  both  from  their  melodious  song  and  con- 
spicuous habits. 

I  here  repeat  a  note  made  about  their  nesting,  which  has 
already  appeared  in  Dresser's  account  of  this  species  : — "  A  pair 
nested  in  a  hole  outside  the  wall  of  my  stable  at  Gibraltar 
in  June  1869.  Five  eggs  were  laid,  which  were  hatched  about 
the  20th.  The  nest,  composed  of  small  dried  bits  of  roots, 
was  very  scanty  and  ill  put  together.  When  the  young  were 
hatched,  I  broke  through  the  wall  from  the  inside  of  the  stable 
to  the  nest,  making  the  hole  large  enough  to  admit  a  small  cage, 
in  which  I  placed  the  nest  and  young ;  and  then  hung  an  old 
coat  over  the  inside  hole,  so  as  to  shut  out  the  light  from  the 
inside,  cutting  a  small  slit  in  the  coat  through  which  to  watch 
the  old  birds  feeding  their  young  within  six  inches  distance. 
Both  birds  fed  them,  at  intervals  of  not  more  than  five  minutes. 
The  food  consisted  almost  entirely  of  centipedes  (Scolopendrce), 
with  now  and  then  a  large  spider  or  bluebottle  fly  by  way  of 
change.  Where  they  could  have  found  so  many  centipedes  it  is 
difficult  to  imagine,  as  they  are  insects  which  lie  hid  all  day 
under  stones  &c.  The  head  was  always  bitten  off,  and  the 
insect  so  mangled  as  to  be  quite  dead.  Two  of  the  five  young 
died  in  the  cage,  from  the  old  birds  not  being  able  to  get  at 
them.  Of  the  other  three,  only  one  attained  maturity,  living 
till  October,  when,  to  my  great  regret,  he  went  the  way  of  all 
pets.  He  was  very  tame,  and  of  most  engaging  habits  and 
disposition — in  fact,  what  the  Spaniards  call  '  simpatico.'  In 
his  early  days  he  was  fed  on  bread  and  bruised  snails  ;  later  on 
he  had  more  fruit,  which  possibly  killed  him. 

"The  Blue  Rock-Thrush  very  often  perches  on  trees,  and  at 
Gibraltar  and  Tangier  is  frequently  seen  on  the  house-tops, 
though  generally  observed  on  bare  rocky  ground.  They  are 
sometimes  found  in  wooded  places,  if  there  are  any  high  rocks  ; 
for  instance,  a  pair  nested  at  the  first  waterfall  near  Algeciraz, 


38  TUKDID.E. 

which  is  in  the  midst  of  a  dense  forest.  They  have  a  habit 
in  the  courting-season  of  flying  straight  out  from  a  rock,  and 
then  suddenly  dropping  with  the  wings  half  shut,  like  a  Wood- 
Pigeon  in  the  nesting-time.  The  Blue  Rock-Thrush  is  very 
fond  of  ivy-berries  and  all  fruit." 

It  seems  that  they  nest  more  than  once  a  year,  as  on  the 
25th  of  April  Mr.  Stark  found  a  nest  with  young  about  a  week 
old,  and  on  the  3rd  of  May  a  nest  with  five  eggs  hard  sat  on, 
the  one  in  my  stable  being  hatched  in  the  end  of  June.  One 
set  of  eggs  obtained  by  Mr.  Stark  were  of  the  usual  delicate 
pale  blue  colour,  but  marked  with  small  russet  spots  at  the  large 
end,  somewhat  like  eggs  of  the  Black  Wheatear  (Saxicola 
leucura). 

All  the  nests  built  on  ledges  of  rocks  and  open  to  view  are 
larger  and  better-built  than  those  placed  in  holes. 

Verner  tells  me  (1894)  that  they  still  occupy  the  same 
situations  on  the  Rock  as  they  did  twenty  years  ago ;  and  he 
knows  no  bird  whose  eggs  are  more  difficult  to  find,  as  during 
time  of  laying  and  incubation  they  seem  to  know  if  they  are 
watched ;  and  the  nests  on  the  Rock  are,  as  a  rule,  more  difficult 
of  access  than  in  the  sierras.  He  found  on  May  22,  1875,  five 
young  able  to  fly ;  May  13,  1877,  five  young  fully  fledged ; 
March  23,  1878,  nest  just  finished ;  April  24,  1879,  nest  just 
completed,  first  egg  on  April  13.  They  never  seem  to  lay  more 
or  less  than  five  eggs  :  this  out  of  many  nests  he  found. 

Male  in  spring.  Uniform  slaty  blue  ;  head  and  neck  with  silvery  gloss  : 
in  autumn  and  winter  with  narrow  whitish  margins  to  the  body-feathers. 
Female.  Browner  than  male.     Length  8£  to  9  inches. 

8.  Monticola  saxatilis,  Linnaeus.    The  White-backed  Rock- 
Thrush. 

"  Is  found  on  passage  only  near  Tangier,  crossing  to  Europe 
in  April.  Is  a  scarce  species,  and  very  rarely  obtained  during 
passage." — Favier. 


SAXICOLA  (ENANTHE.  39 

I  saw  several  near  Tangier  on  the  16th  of  April,  1872,  and  one 
on  the  30th  of  March,  1874,  also  numbers  passing  at  Gibraltar 
on  the  4th  of  April,  1870  ;  and  one  was  seen  there  returning  on 
the  26th  of  September,  1868. 

According  to  Arevalo  this  bird  nests  in  all  the  mountain  ranges 
of  the  Peninsula. 

Tail  chestnut,  two  centre  feathers  darker  than  others  ;  axillaries  and  lower 
parts  chestnut. 

Male.  Head  and  neck  blue  ;  centre  of  back  nearly  white. 
Female.  Brown,  mottled.     Length  7^  inches. 

9.  Saxicola  cenanthe,  Linnaeus.    The  Common  Wheatear. 

Spanish.  Culiblanco,  Ruiblanca :  but  these  names  apply  to  all 
the  Wheatears. 

"  This  is  the  most  common  of  the  '  Traquets,'  except  the 
Stonechat  and  Whinchat,  but  is  only  seen  near  Tangier  on 
migration  in  small  flights  during  March  and  April,  returning  in 
September." — Favier. 

Wheatears  are  abundant  in  Andalucia,  but  seen  only  on 
passage.  First  noticed  on  the  4th  of  March  in  1870  ;  a  single 
male  bird  at  Tangier  on  the  26th  of  March  in  1874  ;  many  seen 
near  Alcala  del  Rio  on  the  4th  of  April  ;  again  passing  in 
numbers  at  Gibraltar  on  the  12th  of  April.  Wheatears  were 
plentiful  near  Casas  Viejas  at  the  end  of  October  and  the  first  part 
of  November,  being  last  seen  on  the  13th  of  that  month.  I  have 
a  note  also  of  observing  six  or  seven  in  the  middle  of  the  Bay  of 
Biscay  on  the  9th  of  October,  when  they  settled  on  the  steamer, 
keeping  with  us  till  night. 

Axillaries  white,  writh  dark  centres.     Rump  white. 

Male.  Above  pale  slate-grey  ;  below  white  ;  ear-coverts  black. 

Female.  Above  dull  brown  ;  below  buff  ;  ear-coverts  dark  brown. 

Young.  Like  the  female,  but  spotted  with  dark  markings  on  feathers 
above  and  below. 

Both  sexes  in  autumn  are  alike,  resembling  female  in  spring,  but  have 
buff  margins  to  all  the  feathers.  Length  6  inches. 


40  TURDID.E. 

10.  Saxicola  stapazina,   Vieillot.     Western   Black  -  throated 
Wheatear. 

"  Passes  near  Tangier  during  March  and  April,  returning  in 
September.  Is  the  most  frequent  after  the  Wheatear,  with 
which  bird  they  travel." — Favier. 

Is  in  Andalucia  apparently  less  common  than  the  Black-eared 
Wheatear,  perhaps  because  they  frequent  higher  ground ;  at 
least  I  have  noticed  them  more  about  mountain-tops.  First 
seen  17th  of  March,  1877  (Verner),  23rd  of  March,  1894 
(Tapatanilla).  They  nest  about  the  same  time  as  the  Black-eared 
Wheatear,  which  they  resemble  in  habits,  nest,  and  eggs. 

Axillaries  black  ;  two-thirds  of  two  centre  tail-feathers  black  with  the 
basal  third  white,  the  others  white  tipped  with  black. 

Male.  Crown,  back,  rump,  breast,  and  belly  white,  suffused  with  buff' 
on  back  and  breast.  Chin  and  upper  throat  black.  Length  5|  inches. 

11.  Saxicola   albicollis,    Vieillot.     The  Black-eared  Wheat- 
ear. 

According  to  Favier,  this  bird  is  less  common  than  Saxicola 
stapazina  near  Tangier,  but  is  met  with  in  the  same  way.  Near 
Gibraltar  they  appeared  to  me  to  be  the  most  frequent,  and  were 
first  seen  there  on  the  3rd  of  April,  1870,  when  several  were 
noticed ;  and  on  the  15th  of  March,  1872,  one  was  observed,  and 
a  single  bird  at  Tangier  on  the  14th.  Seen  at  Tangier  on  the 
18th  March,  1894,  by  Mr.  Irby,  60th  Kifles,  and  by  myself  on 
the  23rd  at  Sierra  Retin.  They  breed  on  the  "  Queen  of  Spain's 
Chair,"  laying  about  the  first  week  in  May,  building  a  loosely 
constructed  nest  among  stones  and  rocks,  very  often  in  the  same 
situations  as  the  Blue  Rock-Thrush.  The  eggs  are  light  blue, 
with  a  zone  of  brown  spots  at  the  large  end. 

Throat  white. 

Male.  Marked  as  last,  but  has  head  and  back  more  whitish,  but  generally 
suffused  with  buff;  lores  and  patch  round  eye  to  side  of  neck  black. 
Female.  Brownish  where  black  in  male.     Length  6^  inches. 


SAXICOLA  LEUCURA.  41 

12.  Saxicola  leucura  (Gmelin).    The  Black  Wheatear. 

Spanish.  Sacristan  (the  Sexton) ;  Pedrero  (the  Stone-mason). 

This  bird  is  merely  named  as  occurring  near  Tangier  by 
Favier,  but  is  found  in  Morocco  in  suitable  localities. 

On  the  Spanish  side  the  Black  Wheatear  is  a  common  and 
conspicuous  bird  at  Gibraltar,  and  to  be  seen  throughout  the 
year;  elsewhere  some  are  migratory,  arriving  in  March,  and 
only  found  on  bare  rocky  ground.  The  nest  is  sometimes  in 
clefts  of  rocks,  so  deep  in  as  to  be  inaccessible. 

Mr.  Stark  took  a  nest  on  the  25th  of  April,  near  Gibraltar, 
containing  four  pale  blue  eggs  hard  sat  on,  marked  with  a  zone 
of  light  reddish-brown  spots.  The  nest  was  very  large,  loosely 
built  with  grass  and  heather-roots,  lined  inside  with  finer  grass, 
two  or  three  feathers  of  the  Neophron,  and  one  bit  of  palmetto 
fibre. 

The  name  of  pedrero  is  applied  to  this  bird  from  their  curious 
habit  of  placing  small  stones  as  a  foundation  to  their  nest,  and 
frequently,  as  when  open  to  view,  making  a  sort  of  wall  or  screen 
of  stones  in  front  of  the  nest. 

Verner  remarks  that  they  much  resemble  the  Blue  Rock-Thrush 
in  habits ;  and  all  the  many  nests  found  by  him  had  foundations 
of  small  stones — the  first  two  had  foundations  only,  but  the  third 
had  a  slight  wall  in  front — and  until  he  heard  from  me  that  this 
bird  was  known  near  Malaga  as  pedrero  he  was  much  puzzled  to 
account  for  the  stones  being  so  placed.  In  some  instances  the 
foundation  appears  to  be  useful  in  keeping  the  nest  dry. 

The  most  remarkable  nest  we  met  with  was  on  the  6th  of 
April,  1894,  in  a  small  hole  in  the  roof  of  a  sandstone  cave  in 
the  Sierra  Bartolome.  This  nest,  which  I  myself  saw  in  situ, 
was  made  of  grass  and  fibres  lined  with  finer  fibres  of  the 
palmetto,  without  wool  or  feathers :  the  cavity  was  filled  by  the 
nest  and  by  the  foundation  and  barrier  of  stones  in  front ;  the 
latter  was  9  inches  long,  the  same  in  width,  and  2^  inches  high. 
Yerner  removed  the  stones  and  found  that  the  wall  in  front 


42 


TURDID^E. 


""C-^— •*-     *•**          *^     ' 

v-<i  ' 


NEST  OF  THE  BLACK  WHEATEAB. 


PRATINCOLA  KUBICOLA.  43 

consisted  of  no  less  than  282  stones,  while  the  'foundation  was 
formed  of  76  more,  a  total  of  358,  the  largest  being  2  inches  long 
by  |  inch  wide  and  J  inch  deep,  weighing  2  oz.,  the  total  weight 
of  all  the  stones  being  4^  Ibs.  A  representation  of  this  nest  is 
given :  the  cave  was  about  six  feet  high  and  nearly  circular,  with 
a  diameter  of  about  twelve  feet. 

Verner  found  the  usual  number  of  eggs  to  be  five,  occasionally 
only  four ;  and  they  appear  to  be  irregular  in  laying,  one  nest 
having  fully  fledged  young  on  the  10th  of  April,  other  birds  not 
laying  till  the  end  of  that  month. 

Male.  General  colour  black  ;  rmnp  and  under  tail-coverts  white. 
Female.  Sooty  brown  where  black  in  male,  and  tinged  with  brown  on 
the  white  parts.     Length  6^  inches. 

13.  Pratincola  rubetra  (Linnseus).    The  WMnchat. 

Moorish.  Erdan  (Earner}.     Spanish.  Zarzalera. 

"  Is  only  a  passing  migrant  near  Tangier,  crossing  to  Europe 
in  April  and  May,  returning  to  winter  further  south  in  September 
and  October.  Is  the  most  common  of  the  Chats,  except  the 
Stonechat." — Favier. 

The  Whinchat  is  met  with  as  above,  on  the  Spanish  side  being 
first  noticed  on  the  7th  of  April ;  on  the  20th,  in  1870,  many 
hundreds  passed  at  Gibraltar,  also  on  the  12th  of  the  same  month 
in  1872,  and  I  noticed  them  as  late  as  the  3rd  of  May.  They 
return  in  September,  being  never  seen  in  winter,  and  are  not  yet 
known  to  breed  in  the  south  of  Spain. 

Male.  Chest  orange-rufous  ;  white  line  over  eye  from  bill  to  nape  ;  basal 
half  of  tail  white,  rest  black. 

Female.  Like  male,  but  paler. 

Young.  Like  old  female,  but  more  spotted,  with  white  tips  to  the 
feathers.  Length  5  inches. 

14.  Pratincola  rubicola  (Linnreus).    The  Stonechat. 
Moorish.  Bou-erdan.     Spanish.  Caganchina,  Cagaropa. 

"  The  Stonechat  is  resident  and  most  abundant  about  Tangier, 


44  TUBDID.E. 

being  seen  in  all  directions,  perched  on  the  tops  of  plants, 
bushes,  and  hedges.  They  nest  from  March  to  July.  Some 
arrive  from  Europe  in  September  and  October,  leaving  in 
February  and  March." — Favier. 

This  bird  is  also  on  the  Spanish  side  one  of  the  most  common 
and  at  the  same  time  conspicuous  birds.  They  increase  perhaps 
in  numbers  in  autumn  and  spring,  but  are  as  common  in  winter 
as  in  summer.  I  found  a  nest  with  five  eggs  hard  sat  on  the 
10th  of  March,  and  have  no  doubt  that  they  breed  more  than 
once  in  the  season. 

Male.  Head,  chin,  and  throat  black  ;  breast  orange-chestnut  ;  upper 
tail-coverts  whitish,  spotted  with  black.  Tail  blackish  brown  ;  outer  weh 
of  outside  feathers  edged  with  buff. 

Female.  Upper  tail-coverts  reddish  brown  ;  lower  parts  rufous. 

Young.  Spotted.     Length  5£  inches. 

15.  Ruticilla  moussieri,  Olph- Gal  Hard.    Moussier's  Redstart. 

Under  the  synonym  of  "  Huticilla  erythrogastra " — a  large 
eastern  Redstart — Favier,  in  his  notes,  has  included  Moussier's 
Redstart  ;  but  his  description  fully  identifies  it  as  a  male 
M.  moussieri.  He  mentions  one  killed  in  1848.  I  obtained 
an  adult  male  at  Tangier  on  the  14th  of  March,  1872,  and 
saw  three  others  killed  in  that  month  in  1874 ;  since  then 
they  appear  to  be  of  irregular  occurrence  in  spring. 

I  never  obtained  one  on  the  Spanish  side  of  the  Straits,  or 
saw  a  Spanish  specimen. 

Male.  Crown,  sides  of  head,  back,  and  wings  black  ;  white  line  over  eye 
from  forehead  to  nape.  White  patch  on  wing.  Rump,  upper  tail-coverts, 
tail  (except  the  two  brown  centre  feathers),  and  underparts  pale  chestnut. 

Female.  Much  as  female  of  Common  Redstart,  but  smaller  and  russet- 
red  below.  Length  4^  inches. 

16.  Ruticilla  phcenicurus  (Linnaeus).    The  Common  Redstart. 
Moorish.  Houmeira  (reddish).     Spanish.  Culirojo. 

"  This  Redstart  is  only  found  on  passage  near  Tangier, 
crossing  the  Straits  in  March  and  April,  returning  in  September 


RUTICILLA  TITYS.  45 

and  October.  It  is  not  so  common  as  Ruticilla  titys,  and  is 
seldom  seen  settled  on  rocks." — Favier. 

The  Common  Redstart  is  seen  in  great  numbers  near  and  at 
Gibraltar  on  passage.  My  earliest  dates  of  arrival  noticed  were 
the  4th  of  April  1868,  5th  of  April  1869,  22nd  of  March  1870, 
28th  of  March  1874  (Tetuan),  23rd  of  March  1894.  In.1872 
they  passed  in  great  quantities  on  the  12th,.  loth,  14th,  and  15th 
of  April,  the  last  noticed  being  on  the  26th,  in  the  Cork-wood. 

They  are  not  known  to  remain  and  nest  near  Gibraltar,  but 
are  reported  by  Arevalo  to  do  so  near  Granada. 

Rurnp  and  tail  chestnut ;  axillaries  chestnut,  or  rufous. 

Male  in  spring.  Black  line  in  front  of  the  white  forehead  ;  chin,  throat, 
and  sides  of  neck  black  ;  back  grey  ;  centre  of  belly  nearly  white. 

Female.  Above  dull  brown  ;  underparts  paler ;  breast  shaded  with 
chestnut  and  white. 

Male  in  autumn  much  as  adult  female. 

Young.  Spotted.     Length  5j  inches. 

17.  Ruticilla  titys  (Scopoli).    The  Black  Redstart. 

Spanish.  Culirojo,  Tintorero. 

"  This  species  is  the  most  common  Redstart  about  Tangier, 
remaining  throughout  the  winter  among  rocks  and  old  buildings. 
They  arrive  during  October,  and  depart  in  March.  The  old 
birds  are  solitary  ;  but  the  immature  birds  keep  together.  They 
shake  their  tails  incessantly,  and,  holding  their  heads  erect,  are 
difficult  to  get  a  shot  at "  (!). — Favier. 

The  Black  Redstart  is  seen  at  Gibraltar,  as  at  Tangier, 
arriving  in  November,  and  never  being  seen  after  March.  They 
nest,  however,  a  little  way  north  of  San  Roque. 

A  specimen  I  killed  at  Gibraltar  had  been  eating  very  small  ants. 

Tail  and  rump  chestnut-red  ;  two  centre  tail-feathers  dark  ;  axillaries 
Hack,  or  greyish  black. 

Male.  Cheeks,  throat,  and  breast  black  ;  primaries  and  secondaries  edged 
on  outer  webs  with  white  ;  centre  of  belly  nearly  white. 

female  and  young  male.  Above  entirely  sooty  brown  ;  underparts  nearly 
as  dark  as  the  upper.  Length  5f  inches. 


46  TURDID.E. 

18.   Ruticilla  cyanecula,  Wolf.     The  White-spotted  Blue- 
throat. 

Spanish.  Soldiya,  Gargantiazul. 

"  Found  near  Tangier  only  on  passage,  and  then  very  rarely. 
I  only  obtained  them  four  times — in  1839,  1844,  1866,  and 
1867.  They  cross  to  Europe  in  February  and  March,  returning 
in  October." — Favier. 

Capt.  Savile  Reid  saw  many  among  reeds  and  rushes  in  the 
marshes  of  Meshree  el  Haddar  in  the  winter  of  1884-85  (Ibis, 
1885,  p.  242). 

The  greater  number  of  this  species  must  pass  further  to  the 
east,  as  about  Gibraltar  I  only  saw  one,  which  I  shot  on  the 
1st  of  March,  as  it  was  perched  on  some  rushes  in  an  old 
"  salina  "  near  Palmones ;  another,  shot  in  November  1873  at 
the  same  place,  was  in  the  possession  of  Capt.  Reid.  They  are 
fairly  common  on  passage  in  the  provinces  of  Seville,  Granada, 
and  Malaga,  but  are  not  known  to  breed. 

This  Bluethroat,  in  1760,  was  called  by  Brisson  the  "  Gorge 
bleue  de  Gibraltar." 

Tail  brown,  with  basal  two-thirds  of  all  but  two  centre  feathers  chestnut. 
Male.  White  spot  in  centre  of  the  blue  throat. 

female  and  young  male.  Below  huffish  white  ;  a  dark  band  across  throat, 
with  occasionally  faint  marks  of  blue  and  chestnut.  Length  6  inches. 

19.  Ruticilla   suecica    (Linnaeus).     The   Red-spotted    Blue- 

throat. 

Recorded  by  Arevalo  from  Malaga  and  Valencia.  This  form, 
which  ranges  furthest  north,  resembles  the  last,  except  in  having 
a  red  instead  of  a  white  spot  on  the  throat.  The  form  with 
entirely  blue  throat  is  known  as  R.  wolfi,  and  has  been  obtained 
at  Malaga. 

20.  Erithacus  rubecula  (Linnaeus).    The  Robin. 

Moorish.  Humar  sidri.     Spanish.  Petirojo. 

"  Is    resident    near    Tangier,   and   very   common  in   all   the 


DAULIAS  LUSCINIA.  47 

gardens  around  the  town.  Numbers  also  migrate,  arriving 
during  October  and  November,  departing  in  February  and 
March." — Favier. 

Common  throughout  Andalucia  in  winter.  The  Robin  only 
comes  to  Gibraltar  from  about  the  middle  of  October  to  the 
middle  of  March,  but  then  in  considerable  numbers.  They  are 
resident  in  the  Cork-wood,  nesting  abundantly  in  April,  where, 
one  day  in  May,  my  attention  was  attracted  by  the  chattering 
and  scolding  of  two  Robins,  evidently  in  a  great  state  of  alarm 
and  excitement ;  close  by  them  was  a  palmetto  bush,  to  and 
from  which  they  were  flying,  hovering  over  it,  but  not  settling. 
At  first  I  thought  a  cat,  or  perhaps  an  ichneumon,  was  lying  up ; 
but  tin  peeping  quietly  into  it,  saw  a  snake,  some  three  feet  long, 
in  the  act  of  swallowing  a  half-fledged  Robin  at  the  edge  of 
its  nest.  I  drew  back  a  pace,  and  fired  a  small  charge  of 
dust-shot  into  the  reptile's  head,  cutting  it  nearly  in  half.  The 
snake,  however,  had  disposed  of  all  the  young  birds ;  so,  though 
too  late  to  save  them,  the  parents  were  rescued,  as  no  doubt  they 
would  have  shared  the  fate  of  their  progeny  but  for  my  inter- 
ference ;  the  snake  I  hung  up  in  the  nearest  bush,  "pour 
encourager  les  autres,"  the  old  Robins  all  the  time  watching  my 
proceedings  ;  and  it  is  to  be  hoped  they  were  able  to  under- 
stand that  their  loss  was  partially  avenged.  The  quantity  of 
young  birds — Robins,  Nightingales,  and  similar  ground-nesting 
Warblers — which  are  destroyed  by  snakes  and  lizards  must 
be  very  great ;  but  many  of  these  birds  in  Andalucia  build 
their  nests  in  bushes  at  some  distance  from  the  ground,  no 
doubt  to  avoid  the  reptiles. 

Chin,  throat,  and  breast  orange- chest  nut. 

Young  in  first  plumage  mottled,  each  feather  orange-buff  tipped  with 
black.  Length  5|  inches. 

21.  Daulias  luscinia  (Linnaeus).    The  Nig-hting-ale. 
Moorish.  Mula  el  hasan  (Owner  of  Beauty,  Favier),  Umm  el 
hasan  (Mother  of  Beauty),     Spanish.  Ruisenor. 


48  TURDJD.E. 

"  This  bird  is  very  common  around  Tangier,  arriving  during 
March  and  April,  passing  on  across  the  Straits  to  return  in 
August  and  September.  Great  quantities  remain  to  breed 
about  the  thick  bushy  places,  chiefly  constructing  their  nest 
with  the  fibres  of  the  palmetto,  the  same  material  used  by  the 
Arabs  in  making  their  tents." — Favier. 

The  Nightingale  is  equally  abundant  on  the  Spanish  side. 
The  number  heard  singing  in  the  Cork-wood  and  elsewhere  is 
perfectly  surprising,  every  clump  of  bramble-brakes  having  its 
pair,  though  in  some  seasons  they  are  more  numerous  than  in 
others  ;  but  there  are  always  a  great  many.  They  are  to  be 
heard  on  the  Rock  for  about  ten  days  or  a  fortnight  after  their 
arrival,  but  nearly  always  pass  on,  though  they  have  been  known 
to  nest,  as  in  1871.  My  earliest  dates  of  their  arrival  are  the 
8th  of  April  1868,  2nd  of  April  1869,  7th  of  April  1870,  1st  of 
April  1871,  21st  of  March  1872,  Tangier;  30th  of  March  1874, 
Tetuan;  noticed  by  Mr.  Meade-Waldo  on  22nd  of  March  1892, 
Tangier;  31st  of  March,  1894,  Sierra  Retin.  Verner's  dates  of 
arrival  are  5th  of  April,  1876  ;  3rd  of  April,  1877  ;  1st  of  April, 
1879.  The  majority  arrive  about  the  12th  of  April. 

They  begin  to  lay  about  the  1st  of  May,  and  usually  build  on 
the  ground  ;  but  sometimes  the  nest  is  placed  in  ivy  or  rubbish 
some  two  or  three  feet  high.  In  swampy  jungles  it  is  built  at 
the  bottom  of  a  bush,  and  has  the  lower  half  constructed  of  dead 
leaves,  the  upper  part  being  made  of  dry  sedges,  like  that  of 
Savi's  Warbler  (Locustella  luscmioides) ;  only  it  is  much  neater, 
and  lined  with  fine  grass,  hair,  and  occasionally  feathers.  As 
a  rule  there  are  young  Nightingales  in  the  Cork-wood  by  the 
24th  of  May.  Verner  found  young  as  late  as  the  3rd  of 
July,  1879. 

General  colour  russet-brown  above ;  upper  tail-coverts  and  tail  chestnut- 
brown  ;  underparts  huffish  white. 

First  primary  longer  than  primary-coverts. 

Young  in  first  plumage  spotted  like  a  Robin.     Length  6|  inches. 


ACCENTOR  COLLAR1S.  49 


Subfamily  ACCENTOEIN.E. 

22.  Accentor  modularis  (Linnseus).    The  Hedge-Sparrow. 

M.  Favier  did  not  include  this  bird  in  his  list  of  Moorish 
birds,  merely  mentioning  it  as  occurring  near  Gibraltar,  having 
met  with  some  during  his  "  triste  sejour "  in  that  place,  in 
November.  I  have  seen  specimens  from  the  African  side  of  the 
Straits.  On  the  Spanish  side  they  are  found  in  winter,  but  not 
commonly,  and  I  have  shot  them  in  the  Cork-wood  in  January ; 
whilst  Arevalo  records  them  as  breeding  near  Malaga  and 
Granada. 

Above  reddish  brown,  streaked  with  blackish  brown  ;  chin,  throat,  and 
breast  bluish  grey  ;  belly  whitish. 

Young.  Streaked  above  and  below ;  no  grey  or  white.  Length  5| 
inches. 

23.  Accentor  collaris  (Scopoli).    The  Alpine  Accentor. 

Spanish.  Serrano  (Mountaineer). 

I  have  only  seen  this  bird  at  the  back  of  the  Rock  at  Gibraltar 
in  winter,  and  shot  one  on  the  1st  of  February,  seeing  others  on 
the  26th  of  the  same  month  in  1870.  Mr.  J.  H.  Gurney,  jun., 
who  was  passing  through  Gibraltar,  was  the  first  to  notice  it  at 
the  signal-station.  Arevalo  records  them  breeding  in  the  sierras 
of  the  Provinces  of  Malaga  and  Granada  ;  and  they  are  no  doubt 
found  on  all  the  high  rocky  ground,  though  I  did  not  meet  with 
any  on  the  Sierra  del  Nino  or  elsewhere. 

Chin  and  throat  dull  white,  with  black  spot  on  each  feather ;  wing- 
coverts  tipped  with  white  ;  tail  dark  brown,  with  pale  buff  tips.  Length 
7  inches. 


E 


50  TURDIDJ2. 

Subfamily  SYLVIIN^E. 

24.  Sylvia  cinerea,  Bechstein.    The  Common  WMtethroat. 

"  Arrives  about  Tangier  and  crosses  to  Europe  in  April  and 
May,  returning  to  winter  further  south  in  September  and 
October.  Is  nearly  as  abundant  as  the  Blackcap,  and  seen  on 
passage  in  small  flights.  On  their  return  they  have  the  top  of 
the  head  the  same  colour  as  the  back,  like  the  females  in 
spring." — Favier. 

Mr.  Meade-Waldo  observed  this  bird's  first  arrival  near  Tangier 
on  the  22nd  of  March,  1894. 

I  never  saw  the  Common  Whitethroat  near  Gibraltar  in 
winter.  Their  first  arrival  was  noticed  in  1870  on  the  7th  of 
April,  in  1871  on  the  7th  of  April,  in  1872  on  the  llth  of  April, 
many  passing  on  the  19th  and  20th,  and  in  1874  on  the  8th  of 
A.pril,  in  1894  on  the  30th  of  March.  They  nest  abundantly  in 
the  Cork-wood,  and  also  in  quantities  in  marshy  places,  building 
their  nests  in  thick  leafy  plants,  often  in  those  of  the  willow- 
herb  (Epilobium) ;  the  average  time  for  their  laying  is,  the  7th  of 
May.  When  looking  for  Savi's  Warbler,  we  sometimes  found 
a  dozen  nests  in  the  day. 

Sack  reddish  broion  •  legs  pale  brown  ;  chin  and  thro.it  much  whiter 
than  the  rest  of  the  underparts  ;  quills  and  wing-coverts  edged  with  pale 
rufous  ;  iris  reddish  broion.  Length  5^  inches. 

25.  Sylvia  curruca  (Linnaeus) .    The  Lesser  Whitethroat. 

This  bird  is,  as  far  as  toy  observations  go,  rare.  A  specimen 
was  obtained  in  my  garden  at  Gibraltar  in  April,  and  another 
on  the  19th  of  April,  1872.  It  does  not  appear  to  have  been 
noticed  by  M.  Favier  near  Tangier. 

Head  and  back  bluish  grey,  back  suffused  with  brown  ;  legs  lead-colour  ; 
iris  yellowish  white.  Length  5£  inches. 


SYLVIA  ATRICAP1LLA.  51 

26.  Sylvia  conspicillata,  Marmora.    The  Spectacled  Warbler. 

Favier  merely  says  of  the  Spectacled  Warbler,  that  about 
Tangier  they  are  not  common,  and  only  seen  on  passage  north 
in  March.  He  gives  no  date  of  their  autumnal  migration,  but 
states  that  they  pass  the  winter  somewhere  further  south. 

The  earliest  date  on  which  they  were  obtained  near  Gibraltar 
was  on  the  10th  of  March.  Is  a  conspicuous,  scrub-haunting 
bird,  frequenting  dry  and  more  open  ground  than  the  White- 
throat,  often  being  seen  among  cactus  bushes.  A  sure  place  for 
finding  them  is  on  the  Carteian  hills.  They  remain  during  the 
breeding-season,  but  I  did  not  myself  find  the  nest. 

Chin  and  upper  throat  white,  axillaries  rosy  rufous. 
Male.  Brown-grey;  eyelid  red  ;  secondaries  and  wing-coverts  broadly 
edged  with  chestnut ;  below  pale  rosy  rufous  ;  centre  of  belly  white. 
Female.  Duller  in  plumage.     Length  8^  inches. 
Irides  very  light  brown  ;  inside  of  mouth  pale  yellow. 

27.  Sylvia  hortensis,  Bechstein.    The  Garden- Warbler. 
"Found  near  Tangier,   on  passage  to  Europe,  in  April  and 

May,  returning  in  October,  when  it  is  nearly  as  plentiful  as  the 
Common  Whitethroat." — Favier. 

The  Garden- Warbler  mostly  arrives  during  the  middle  of 
April,  and  was  first  observed  on  the  10th.  The  latest  I  saw  was 
on  the  7th  of  October.  They  nest  around  Tangier  and  in  the 
Cork-wood,  laying  about  the  10th  of  May,  and  are  brought  into 
the  market  at  Gibraltar  as  "  becafigos  ;  "  for  later  in  the  season, 
like  most  of  the  family,  they  are  great  devourers  of  figs. 

General  colour  above  and  tail  olive-brown  ;  underparts  greyish  white. 
Head  same  colour  as  back ;  axillaries  and  under  wing-coverts  buf ;  chin 
and  throat  not  whiter  than  rest  of  underparts.  Length  5-£  inches. 

28.  Sylvia  atricapilla  (Linnaeus).    The  Blackcap. 
Moorish.  Shorrir  el  Kebir. 

"  Is  nearly  as  common  as  S.  melanocephala  about  Tangier, 

E2 


52  TUKD1D.E. 

being  seen  on  all  sides  during  migration,  passing  north  in 
January  and  February,  returning  in  October.  Many  remain  to 
nest." — Favier. 

The  Blackcap  is  to  be  seen  during  every  month  in  the  year, 
but  is,  of  course,  most  common  in  February  and  October.  They 
sometimes  nest  on  the  Rock,  always  plentifully  in  the  Cork-wood. 
The  young  have  been  seen  fully  able  to  fly  on  the  24th  of  May. 
Verner  found  a  nest  with  three  eggs  on  the  15th  of  March,  1877, 
in  the  Convent  Garden,  and  a  nest  with  four  hard-set  eggs  on  the 
16th  of  March,  1880,  in  the  Mount  Garden.  They  chiefly  fed 
in  my  garden  for  some  time  on  the  seed  of  the  so-called  "  pepper- 
tree  "  (Schinus  molle),  in  company  with  Black-headed  Warblers, 
and,  to  my  surprise,  with  the  Black  Redstart ;  at  least  the  latter 
were  seen  pecking  at  the  seeds. 

The  species  may  be  distinguished  at  a  glance  from  the  other 
black-headed  Warblers  met  with  near  Gibraltar  by  the  absence 
of  white  on  the  tail. 

General  colour  above  ashy  brown  ;  chin  and  throat  bluish  grey  ;  tail  without 
any  white. 

Male.  Head  black.     Length  5|  inches. 
Female.  Head  rufous.     Length  6£  inches. 

29.  Sylvia    orphea,    Temmiuck.        The    Western    Orphean 
Warbler. 

Andalucian.   Canaria. 

"This  Warbler  passes  by  Tangier  in  April  and  May  to  return 
in  September,  travelling  in  company  with  the  Whitethroats. 
Is  not  common,  and  in  some  years  scarcely  met  with." — Favier. 

The  Orphean  Warbler  did  not  appear  to  be  common  near 
Gibraltar,  and  I  could  not  get  any  till  the  17th  of  May,  1871, 
when  I  found  a  nest  on  a  branch  of  a  pine-tree  in  the  "  Second 
Pine-wood,"  and  shot  both  the  old  birds,  which  were  very  fearless, 
especially  the  female,  who  contained  an  egg  ready  for  exclusion. 
The  nest  had  only  three  eggs  in  it,  and  was  badly  built,  being 
composed  of  grass  and  lichens.  Verner  found  this  Warbler  on 


SYLVIA  SUBALPINA.  53 

8th  of  May,  1875,  breeding  in  the  cork-woods  along  the  base  of 
the  Retin  Hills ;  the  nests  were  on  the  tops  of  branches  of  small 
cork-trees,  about  ten  feet  from  the  ground. 

It  would  seem  that  this  Warbler  must  chiefly  pass  further  to 
the  east.  They  nest  around  Seville  and  are  common  about 
Madrid ;  but  I  did  not  ascertain  that  they  bred  near  Tangier. 

Male.  General  colour  above  ashy  grey ;  chin,  throat,  and  underparts 
white  ;  outer  pair  of  tail-feathers  with  outer  web  white,  inner  web  brown ; 
next  pair  white  only  at  tips  ;  head  black  ;  legs  bluish ;  irides  yellow. 

Female.  Like  male,  but  the  head  only  slightly  darker  than  back. 
Length  6 1  inches. 


30.  Sylvia  subalpina,  Bonelli.     The  Subalpine  or  Chestnut- 
breasted  Warbler. 

Favier  merely  states  that  "  this  species  occurs  near  Tangier  on 
passage  in  March  and  April,  and  again  in  October." 

They  are  not  often  noticed  near  Gibraltar ;  but  I  shot  one  on 
the  20th  March,  1870,  and  at  Tangier  on  the  26th  of  March  and 
27th  of  April,  1874.  Mr.  L.  P.  Irby  obtained  one  at  Tangier  on 
the  18th  March,  1894.  On  the  27th  of  March,  1871,  I  saw 
eight  or  ten  among  the  flowers  and  trees  on  the  Alameda  de 
Apodaca  at  Cadiz ;  they  were  exceedingly  tame,  and  I  watched 
them  for  a  long  time  hopping  about  in  and  out  among  the 
flowers  like  a  common  Wren.  One  or  two  were  very  bright- 
coloured  males.  We  also  saw  this  Warbler  on  the  25th  of  April, 
1869,  in  the  Goto  del  Key.  Lord  Lilford  informs  me  he  found  a 
nest  early  in  May,  built  in  a  gum-cistus  bush  in  the  Goto  del 
Donana,  the  eggs  being  very  hard  sat-on. 

I  never  had  the  good  fortune  to  discover  a  nest ;  but  they 
build  much  in  the  same  manner  and  situations  as  S.  melano- 
cephala,  the  eggs  being  also  very  similar  to  those  of  that  bird. 

Owing  to  the  chestnut  colour  of  the  throat  and  breast  of 
the  male,  this  species  is  very  apt  to  be  confounded,  when  flying 
and  hopping  about,  with  the  Dartford  Warbler,  but  the  length  of 


54  TTJRDIDJ3. 

the  tail,  shorter  than  the  wing,  easily  distinguishes  them  from 
that  bird. 

Male.  Above  slaty  grey ;  chin,  throat,  and  breast  chestnut;  narrow  white 
line  from  base  of  bill  to  shoulder. 

Female.  Above  dull  brown;  buffy  white  where  chestnut  in  the  male. 
Eyelids  red  in  both  sexes.  Length  5  inches. 

31.  Sylvia  melanocephala    (Gmelin).       The    Black  -  headed 
Warbler. 

Moorish.  Shorrir  (Favier}.     Spanish.  Palmera. 

"  This  Warbler  is  resident  and  very  abundant  near  Tangier ; 
some  migrate,  crossing  the  Straits  during  February  and  March, 
returning  in  September.  They  are  to  be  seen  everywhere, 
nesting  in  small  thorny  bushes.  The  nest  is  not  well  built,  and 
is  made  of  strips  of  plants  and  blades  of  grass,  without  roots ; 
rarely  there  is  a  little  wool.  It  is  lined  with  the  down  of  some 
cotton-like  plant,  fine  fibres  of  roots,  and  a  few  horse-hairs." 
They  lay  from  April  to  July." — Favier. 

The  Black-headed  Warbler,  equally  common  around  Gibraltar, 
is  found  in  all  scrub,  gardens,  and  in  the  midst  of  woods,  scolding 
with  a  chattering  noise  much  like  that  of  our  common  Wren. 
They  might  well  be  named  the  Gibraltar  Warbler,  being  almost 
the  only  species  which  is  a  regular  resident  on  the  Rock.  In 
habits  they  much  resemble  the  Blackcap,  but  are  more  restless 
and  obtrusive,  and  consequently  more  conspicuous  ;  the  contrast 
between  the  jet-black  head  of  the  adult  male  and  the  white 
throat  also  renders  them  more  liable  to  be  noticed.  There  were 
in  different  years  several  nests  in  my  garden,  which  were  carefully 
preserved  ;  but  what  with  cats  and  inquisitive  human  beings, 
they  seldom  succeeded  in  rearing  their  young.  The  earliest  egg 
laid  was  on  the  12th  of  March ;  this  was  built  in  a  small  rose- 
bush, and  was  spoiled  by  a  gale  of  wind,  which  blew  all  the  eggs 
out  of  it,  being  the  only  one  I  ever  saw  in  what  could  be  called 
an  open  bush.  All  the  others  were  placed  in  thick  bushes, 
generally  box,  about  two  to  four  feet  from  the  ground,  and 


SYLVIA  UNDATA.  55 

were  formed  of  grass  with  a  few  bits  of  cotton-thread,  lined  with 
hair ;  but  they  also  nest  at'  some  height  on  boughs  of  trees.  The 
eggs  vary  in  number  from  three  to  five.  The  male  assists  in 
incubation. 

This  bird  is,  like  the  Blackcap  and  Garden-Warbler,  very  fond 
of  figs  and  grapes  and  all  kinds  of  fruit.  The  feathers  at  the 
base  of  the  bill  and  the  throat  are  often  much  coloured  with  the 
pollen  of  cactus,  aloe,  and  other  flowers,  also  with  the  seed  of  the 
"  pepper- trees." 

Tail  longer  than  wing  :  claws  and  throat  pure  white  ;  eyelids  brick-red. 
Male.  General  colour  above  slaty  grey  ;  nape  and  head  black  ;  below 
greyish  white. 

Female.  Above  brownish  ;  head  nearly  black.     Length  4^  inches. 


32.   Sylvia   undata    (BodJaert).       The   Dartford    or    Furze 
Warbler. 

Spanish.  Colorin,  Caganchina. 

"  Is  resident  but  not  abundant  near  Tangier.  Some  migrate 
to  Europe  in  March,  to  return  in  August.  It  is  solitary 
in  habits.  They  make  a  clumsy  nest  of  grass  and  roots, 
lined  with  very  fine  coils  of  palmetto-fibre,  laying  in  April." 
— Famer. 

The  Dartford  Warbler  is  resident  and  not  uncommon  in  all 
the  scrub-covered  hills  on  the  coast  near  Gibraltar,  particularly 
about  San  Roque,  but  is  most  abundant  on  the  sides  of  the 
sierras,  nesting  in  the  heather  about  the  8th  of  April,  on  which 
date  Mr.  Stark  found  a  nest  near  Algeciraz  with  three  eggs. 
There  is  no  doubt  they  nest  at  Gibraltar,  as  they  occasionally 
remain  there  through  the  summer. 

Upper  parts  blackish  brown  ;  throat,  breast,  and  sides  cliestnut-brown  ; 
feathers  of  chin  and  throat  tipped  with  white  ;  tail  long  and  graduated, 
half  the  length  of  the  bird  ;  eyelids  arid  iris  red  in  adults,  yellow  in  young. 
Length  5  inches. 


56  TURDID^E. 

33.  Aedon   g-alactodes   (Temminck).     The   Western   Rufous 
Warbler. 

Moorish.  Houmeira.     Spanish.  Alzacola,  Eubita,  Vinadera. 

"Abundant  in  the  vicinity  of  Tangier,  arriving  in  April  and 
May,  returning  during  September,  many  remaining  to  breed. 
Their  habits  are  the  same  as  those  of  the  Nightingale.  The 
nest,  large  and  well  built,  is  placed  at  some  height  from  the 
ground,  in  thick  foliage.  The  eggs,  from  five  to  six  in  number, 
only  differ  from  Sparrow's  eggs  in  the  spots  being  more  reddish. 
The  males  assist  in  incubation." — Favier. 

On  the  Spanish  side,  this  "  Cocktail "  Warbler,  as  they  may 
well  be  called  from  their  well-known  habit  of  continually  jerking 
their  tails  up,  is  very  plentiful,  frequenting  sandy  lanes  hedged 
with  aloes  and  prickly  pears,  such  as  those  close  to  the  First 
Yenta,  near  Gibraltar.  As  Favier  remarks,  they  resemble 
the  Nightingale  very  much  in  some  of  their  habits,  and  are 
at  first  sight  very  likely  to  be  mistaken  for  that  bird  ; 
only  the  Nightingale  comes  some  three  weeks  or  a  month 
earlier. 

The  Rufous  Warblers  mostly  arrive  near  Gibraltar  between  the 
1st  and  5th  of  May.  The  earliest  I  noticed  in  1869  was  on  the 
28th  of  April,  in  1870  on  the  29th,  in  1871  on  the  22ud,  and  in 
1 872  on  the  28th  of  that  month,  the  migration  lasting  for  weeks. 
They  nest  about  the  last  fortnight  in  May. 

In  places  where  there  are  many  vineyards  (which  they 
frequent)  they  are  known  as  Vinadera.  Alzacola  is  the  local 
name  about  Gibraltar ;  and  "  Cocktail "  is  very  nearly  a 
translation  of  it.  "  Eufous  Sedge- Warbler,"  as  this  bird  has 
been  called,  is  most  inappropriate,  as  they  are  never  seen  near 
either  water  or  sedges. 

Upper  parts  pale  chestnut-brown.  Tail  long  and  graduated  ;  the  two 
centre  feathers  rich  reddish  buff ;  the  others  reddish  buff,  except  the  ends, 
which  are  black,  tipped  with  white.  Length  7  inches. 


ACROCEPHALUS  STREPERUS.  57 

34.  Acrocephalus   turdoides    (Meyer).     The    Great    Reed- 

Warbler. 

Spanish.  Carrizalero ;  from  "  carrizo,"  reed-grass  (Arundo 
pJiragmites). 

This  large  species,  though  occurring  in  Morocco,  is  not 
included  in  Favier's  list. 

They  are  exceedingly  plentiful  in  Andalucia,  arriving  in  April, 
and  chiefly  frequenting  tall  reed-beds :  they  are  very  noisy,  and,  like 
other  aquatic  Warblers,  conceal  themselves  at  the  slightest  alarm. 
However,  by  ensconcing  one's  self  and  remaining  quiet  among 
the  rushes,  they  are  easily  obtained,  as  they  soon  come  out  and 
sit  singing  and  chattering  on  the  top  of  some  tall  reed.  They 
nest  late  in  May,  and  build  a  nest  interwoven  with  and  suspended 
between  reeds,  resembling  that  of  the  Reed- Warbler  (Acro- 
cephalus streperus),  only  of  course  much  larger.  I  have  seen 
them  building  during  the  first  fortnight  in  May,  picking  and 
carrying  away  the  down  of  the  "  bulrush  "  (Typha  latifolia)  to 
use  in  constructing  their  nests. 

Inside  of  mouth  orange-yellow.  Upper  parts  and  tail  olive-brown, 
below  pale  buff ;  centre  of  belly  whitish  ;  tail  graduated  and  long  ;  legs 
pale  brown.  Length  8  inches. 

35.  Acrocephalus  streperus  (Vieillot).    The  Reed- Warbler. 

Not  mentioned  by  Favier  as  occurring  in  Morocco ;  nor  doe 
Mr.  Drake  appear  to  have  observed  any  in  that  country. 

On  the  Spanish  side  they  appear  in  spring.  The  exact  date  of 
arrival  I  could  not  ascertain  ;  but  it  is  somewhere  about  the  end 
of  March.  I  never  met  with  the  Reed- Warbler  during  the  winter 
months ;  but  in  the  marshes  at  Casas  Viejas,  about  the  first 
week  in  May,  we  found  them  breeding  in  abundance.  They 
keep  among  the  sallow-bushes,  but  build  their  beautiful  nests 
suspended  on  the  dead  stems  of  the  Epilobium  Jiirsutum  or 
willow-herb,  which  grows  in  luxuriant  tufts  in  the  swampy 


58  TUKDID^E. 

jungle.  These  nests  are  constructed  externally  of  strips  of  the 
rind  or  peel  of  the  dead  Jfyilobium-stems  interwoven  with  sallow- 
cotton,  the  interior  being  composed  of  fine  grass  lined  with  the 
same  material.  The  usual  number  of  eggs  was  four,  of  a  pale 
greenish  colour,  marked  all  over  with  ashy  spots.  One  nest  we 
found  contained  a  Cuckoo's  egg. 

Upper  parts  olive-brown  ;  below  pale  buff ;  both  distinctly  diffused  with 
rufous,  most  so  on  rump  and  upper  tail-coverts  ;  legs  and  feet  slaty  brown. 
Length  5^  inches. 


[Acroceplialus  palustris  (Bechstein).  The  Marsh-Warbler  is  said  to 
occur  near  Granada,  but  I  have  never  seen  a  specimen  :  it  nests  in  drier 
places  than  the  Keed-Warbler,  though  near  water  ;  the  song  and  eggs  also 
differ  from  those  of  that  bird,  which  it  resembles,  except  having  the  upper 
parts  olive-greenish  without  any  rufous  tinge  ;  legs  and  feet  pale  brown.] 


36.    Acrocephalus    phragmitis    (Bechstein).      The    Sedge- 
Warbler. 

Favier  says  this  is  a  very  rare  species  near  Tangier,  that  he 
seldom  saw  more  than  one  or  two  on  passage,  in  March  or  in 
September. 

On  the  Spanish  side  I  only  noticed  one  in  April  1894;  but 
this  common  English  migratory  bird  is  reported  from  Malaga 
and  Granada. 

Upper  parts  brown,  each  feather  having  a  dark  centre  ;  crown  blackish 
brown,  streaked  with  lighter  brown  ;  broad  buffish-white  eye-stripe. 

Young.  More  yellow  ;  a  few  dusky  streaks  on  chest.  Length  4|-5 
inches. 


37.   Acrocephalus   aquations  (J.  F.  Gmelin).     The  Aquatic 
Warbler. 

This  species,  mentioned  by  Mr.  Drake  as  having  been  met 
with  in  Morocco,  is  found  in  Andalucia,  breeding  near  Malaga  ; 
but  I  only  saw  it  once  near  Gibraltar.  Verner  found  a  nest,  in 


LOCUSTELLA  LUSCINIOIDES.  59 

May  1875,  near  Casas  Viejas,  built  on  the  ground  amidst  the 


Much  as  last,  but  is  yellower  in  tint  of  plumnge  and  has  two  hroad  dark 
bands  along  the  crown,  with  conspicuous  pale  line  down  the  centre  of 
crown. 


38.  Locustella  naevia  (Boddaert).    The  Grasshopper  "Warbler. 

Is  recorded  from  Morocco  and  also  from  Malaga  in  winter ; 
but  I  did  not  observe  any  near  Gibraltar. 

Above  olive-brown,  obscurely  spotted  on  back,  "  like  a  Lark."     Ttdl 
brown,  indistinctly  marked  with  transverse  bars  of  darker  brown. 
Young.  Feathers  on  throat  with  dark  centres.     Length  5^  inches. 

39.  Locustella  luscinioides  (Savi).    Savi's  Warbler. 

Recorded  by  Mr.  Drake  as  met  with  in  Morocco. 

In  Andalucia,  I  only  found  them  in  one  situation  where  once, 
when  snipe-shooting  in  winter,  having  noticed  some  old  nests  in 
the  sedges  which  apparently  belonged  to  this  species,  I  made 
up  my  mind  to  try  the  next  spring  for  them.  However,  for  two 
years  I  was  unable  to  do  so ;  but  in  1874  I  went  to  this  place 
in  May  with  two  friends,  Mr.  Stark  being  one,  and  we  succeeded 
in  finding  thirteen  nests. 

The  first  nest  was  found  by  Mr.  Denison,  on  the  4th  of  May, 
and  contained  four  fresh  eggs  ;  the  others  as  follows : — on  the 
6th,  one  nest  with  four  fresh  eggs  ;  on  the  7th,  three  nests — one 
empty  (deserted),  two  with  four  eggs  each,  one  lot  fresh,  the 
other  hard  sat-on ;  on  the  8th,  one  nest  procured  with  three 
eggs  slightly  sat-on,  and  one  nest  with  five  fresh  eggs ;  on  the 
9th,  two  nests  with  four  eggs  each,  all  hard  sat-on,  and  one  nest 
with  three  young  fully  fledged  ;  on  the  llth,  one  nest  with  five 
fresh  eggs  ;  and  on  the  13th,  one  nest  with  two  fresh  eggs. 

By  this  it  will  be  seen  that  the  time  of  their  breeding  is  rather 
variable.  I  do  not  like  to  give  the  name  of  the  exact  locality 


CO  TURDID.E. 

where   these    birds   nest,    as    Savi's    Warbler    might   suffer   in 
consequence. 

The  precise  time  of  their  arrival  I  could  not  ascertain ;  but  it 
is  about  the  1st  of  April ;  and  they  are  all  gone  by  September. 
The  nests,    sometimes   very   near   to    one   another,   were   most 
difficult  to  find,  and,  without  exception,  built  in  places  where 
the  mud  and  water  varied  in  depth  from  two  or  three  inches  to 
perhaps  two  feet.     All  but  one  were  in  sedges,  so  well  concealed 
as  only  to  be  found  by  accident.     We  spent  sometimes  the  whole 
day  in  these  marshes,  looking  in  vain,  with  a  gun  in  one  hand 
and  a  sickle  in  the  other,  the  latter  used  to  open  the  sedges  with, 
as  it  cut  one's  fingers  severely  to  try  and  move  them  with  the 
hand.     What  with  the  hot  sun  and  the  stink  of  the  mud,  we 
used   to   despair   utterly   after   hours    of   fruitless    search,    but 
generally  found  a  nest  in  the  evening.     The  whole  marsh  was 
trodden  down  by  us  as  if  a  herd  of  cattle  had  been  wandering 
about ;  but  perhaps  the  next  day,   going  over  the  same  ground, 
we  would  find  a  nest  in  a  bunch  of  sedges  which  had  been  passed 
by  within  a  yard.     The  nests  were  all  alike,  loosely  and  clumsily 
built,  solely  constructed  of  dead  sedge,  often  placed  so  close  to 
the  water  that  the  base  was  wet ;  they  were  always  in  the  open 
marsh,  none,  that  we  saw,  under  bushes  or  in  tall  rushes  or  reeds, 
and  the  single  nest  that  was  not  in  sedges  was  in  a  tuft  of  the 
spiky  rush  so  common  in  wet  ground.     In  this  case  (the  first 
one,  found  by  Mr.  Denison)  the  bird  flew  off — the  only  instance 
in  which  it  did  so,  as  they  creep  off  generally  like  a  mouse.     On 
one  occasion  I  cut  away  all  the  sedge  round  the  nest,  except 
just  the  patch  in  which  it  was  built,  as  I  wanted  to  shoot  the 
bird  from  the  nest  to  make  certain  of  the  identity  of  the  eggs  ; 
but  even  then,  after  watching  the  old  bird  go  in  to  the  nest  she 
would  not  fly  off,  but  ran  across  the  open  space  which  had  been 
cut  away  till  she  gained  the  shelter  of  the  uncut  sedges.     Much 
more  frequently  seen  than  Cetti's  Warbler,  the  great  difficulty 
is  in  finding  them  when  shot.     If  killed  on  the  wing,  it  is  almost 


CETT1A  CETTII.  61 

hopeless  to  look  for  them  ;  and  those  that  I  did  obtain  I  had  to 
thank  my  dog  for  finding,  though  he  spoilt  one  or  two.  They  are 
most  easily  to  be  got  in  the  morning  and  evening,  when  the  male 
perches  on  a  sallow  bush  or  tall  reed  and  sings  his  grasshopper- 
like  song,  or  rather  whir. 

We  only  found  them  in  one  particular  locality:  in  other 
marshes,  very  similar  in  appearance,  we  failed  to  hear  or  see 
them ;  and  they  probably  require  a  very  large  extent  of  sedge 
(Cladium  mariscus). 

The  eggs  are  of  a  whitish  ground-colour,  marked  all  over  with 
minute  spots  of  brown,  thicker  at  the  larger  end,  often  forming 
a  well-marked  zone.  Sometimes  the  ground-colour  is  buff;  but 
there  were  only  two  or  three  of  this  hue. 

Savi's  Warbler  has  been  obtained  near  Seville  by  Ruiz,  and 
Arevalo  records  them  on  passage  near  Malaga. 

Above  uniform  russet-brown  ;  chin  whitish  ;  underparts  light  brown. 
Tail  slightly  cuneiform,  indistinctly  barred  with  dark  brown. 

Legs  and  feet  pale  brown,  the  claws  darker  ;  irides  olive-brown  ;  lower 
mandible  dirty  white,  with  dusky  tip  ;  upper  one  blackish  ;  inside  of  mouth 
of  adult  pale  salmon-colour,  that  of  young  bright  yellow.  Length  6 
inches. 

40.  Lusciniola  melanopog-on  (Temminck).     The  Moustached 

Swamp-Warbler. 

Has  been  obtained  near  Malaga  ;  is  common  and  breeds  near 
Valencia.  Inhabits  rushy  and  reedy  swamps. 

Marked  on  back  much  as  Sedge-Warbler :  crown  nearly  black,  broad 
white  eyebrow,  and  blackish  stripe  through  eye.  First  primary  half  the 
length  of  the  second.  Length  5j  inches. 

41.  Cettia  cettii  (Marmora).    Cetti's  Bush- Warbler. 

Favier  states  that  "  this  Warbler  is  rare  near  Tangier,  and 
seen  on  passage  in  February  and  March,  to  return  in  October." 
This,  however,  is  quite  different  from  my  own  observations. 
They  certainly  are  not  rare  in  spring  near  Tangier,  where,  as 


62  TURDID.E. 

on  the  Spanish  side,  wherever  there  are  thick  bushes  (generally 
bramble-brakes  close  to  water)  Cetti's  Warbler  is  to  be  heard. 
Perhaps  many  migrate ;  but  at  Casas  Viejas  they  are  quite  as 
common  during  the  winter  months  as  at  any  other  season,  and, 
somewhat  like  our  own  Robin,  may  be  heard  singing  at  all  times. 
Very  difficult  to  see  in  the  breeding-season,  in  the  winter  months 
they  do  not  skulk  so  much.  They  are  excessively  restless,  being 
ever  on  the  move ;  and  often  in  the  winter,  when  hidden  up  in 
the  sotos  near  Casas  Yiejas,  have  I  watched  them  quite  close  to 
me ;  but  the  slightest  movement  on  my  part  sent  them  off  to 
the  thickest  depths  of  the  jungle.  In  the  breeding-season  it  is 
almost  impossible  to  catch  a  glimpse  of  one. 

The  only  chance  of  shooting  them  is  at  the  nest,  which  is 
always  placed  some  distance  from  the  ground,  generally  at  a 
height  of  about  two  or  three  feet,  and  is  either  situated  in  a 
thick  bush  or  (when  in  a  bushy  swamp)  constructed,  somewhat 
like  the  nest  of  the  Reed- Warbler,  on  the  stalks  of  reeds  and 
Epilolium.  These  nests,  extremely  difficult  to  find,  are  built  of 
bits  of  small  sedges,  intermingled  with  willow-cotton,  and  chiefly 
lined  outside  with  strips  of  the  stems  of  the  Epilolnum,  inside 
with  fine  grass,  a  few  hairs,  and  bits  of  cotton  at  the  top.  Those 
nests  built  in  bushes  are  chiefly  constructed  with  grass  and 
cotton,  and  are  entirely  lined  with  hair.  All  the  nests  are  deep 
and  cup-shaped,  largest  at  the  base,  measuring  about  4J  inches 
in  height,  the  inside  depth  being  2J-,  the  internal  diameter 
2f  inches.  The  beautiful  pink  eggs,  which  lose  much  of  their 
beauty  when  blown,  are  laid  about  the  end  of  April,  and  are 
usually  five  in  number ;  but  I  have  known  only  three. 

The  birds  are  rather  irregular  as  to  the  time  of  nesting,  as  I 
have  seen  nests  nearly  on  the  point  of  hatching  and  others  with 
fresh  eggs  on  the  same  day  (13th  of  May). 

Tail  much  graduated  and  of  ten  feathers,  with  indistinct  bands  of  dark 
brown. 

Sexes  alike,  males  slightly  larger. 

Head  and  upper  parts  rusty  brown.  Chin,  throat,  and  centre  of  belly 
white.  Inside  of  mouth  yellowish.  Length  5£  inches. 


PHYLLOSCOBUS- RUFUS.  63 

42.  Phylloscopus  sibilatrix  (Bechstein).    The  Wood- Warbler. 
This  species  is  not  mentioned  by  Favier  as   occurring  near 

Tangier,  where,  however,  they  are  found,  though  not  commonly. 
On  the  Spanish  side  of  the  Straits  it  is  the  scarcest  of  the  four 
species  of  Phylloscopus,  and  was  first  seen  on  the  22nd  of  April. 
I  killed  one  in  my  garden  at  Gibraltar,  and  some  remain  during 
the  nesting-season  in  the  Cork-wood,  where  Verner  found  a  nest 
on  the  25th  of  June  ;  but  we  did  not  observe  them  in  winter,  and 
were  unable  to  notice  the  date  of  their  departure  south. 

Nest  domed,  but  not  lined  with  feathers,  like  those  of 
P.  rufus  and  P.  trochilus,  and  is  always  placed  on  the  ground. 

The  bird  is  easily  distinguished  by  the  streak  of  bright  yellow 
over  the  eye  and  the  white  colour  of  the  underparts,  and  is  the 
largest  species  of  the  genus  found  near  Gibraltar. 

Distinct  light  streak  from  base  of  bill  over  eye  to  crown  ;  white  belly, 
contrasting  with  yellow  breast  and  throat.  Length  5j  inches. 

43.  Phylloscopus    bonellii    (Vieillot):       Bonelli's     Willow- 

Warbler. 

"  Found  during  migration  near  Tangier,  in  company  with 
P.  rufus  and  P.  trochilus,~but  is  not  so  numerous.  They  return 
in  September." — Favier. 

This  species,  about  the  size  of  the  Chiffchaff,  is  found  in 
plenty  near  Gibraltar,  nesting  in  the  fern  in  the  Cork-wood ;  the 
earliest  I  noticed  arriving  was  on  the  1st  of  April.  Verner,  on 
2nd  May,  1877,  found  a  nest  in  a  genista-bush  about  three  feet 
from  the  ground. 

They  are  not  found  in  winter. 

Head  and  back  olive-brown,  with  green  tinge;  rump  yellow;  breast  and 
belly  white  ;  chin  and  throat  tinged  with  yellow  ;  eye-streak  greyish  white 
Length  4|  inches. 

44.  Phylloscopus  rufus  (Bechstein).    The  Chiffchaff, 

Spanish.  Mosquilla. 

According   to   Favier  this  bird  is  nearly  as  common  as   the 


TURDID.E. 

Willow- Warbler,  crossing  to  Europe  in  February,  March,  and 
April,  returning  in  October  and  November. 

The  ChiffchaiF  is  to  be  seen  throughout  the  year  in  the  Cork- 
wood, but  is  most  common  from  November  to  March.  I  found  a 
nest  on  the  21st  April  in  a  bush  about  six  inches  from  the  ground. 

The  four  Spanish  species  of  Phylloscopus  all  build  domed 
nests,  usually  on  the  ground,  but  occasionally  in  bushes  or  fern  at 
an  elevation  of  sometimes  two  feet  or  more  above  the  ground; 
this  is  particularly  the  case  with  Bonelli's  Willow- Warbler. 

The  Chiffchaff  is  difficult  to  tell  from  the  common  Willow- 
Warbler,  but  is  always  smaller  and  the  legs  are  darker,  being 
almost  black,  the  eyebrow  is  not  so  well  defined,  and  it  is  a  more 
dull-coloured  bird  than  P.  trocliilus.  The  note  is  also  very 
different  and  distinct.  The  young  of  all  the  species  are  more 
highly  coloured  than  the  adult  birds ;  but  the  genus  is  a  very 
troublesome  and  perplexing  one  to  the  student,  and  only  to  be 
elucidated  by  observing  the  different  species  in  a  wild  state,  and 
listening  to  their  notes.  The  skins  shrink  and  the  colours  fade 
so  much  that  a  table-naturalist  is  much  the  most  puzzled  by  them. 

Above  olive-green,  below  yellowish  white.  Smaller  than  next,  and  duller 
in  plumage.  Legs  nearly  black.  Length  4|  inches. 

45.  Phylloscopus  trochilus  (Linnaeus).    The  Willow-Warbler. 

Moorish.  Simriz.     Spanish.  Mosquilla. 

"  The  most  common  of  Willow- Wrens  near  Tangier ;  crosses 
the  Straits  in  April,  returning  in  November." — Favier. 

There  is  no  doubt,  although  I  did  not  find  a  nest,  that  this 
species  breeds  near  Tangier.  In  the  vicinity  of  Gibraltar  they 
are  to  be  found  throughout  the  year  in  the  Cork-wood,  where 
they  breed,  and  I  have  seen  the  young  able  to  fly  on  the  8th 
of  May.  Although  universally  distributed  in  winter,  they  are 
most  common  when  on  passage  in  March  and  October. 

Much  as  last,  but  larger  and  brighter.  Legs  light  Irown.  Length  5 
inches. 


HYPOLAIS  OPACA.  65 

46.  Hypolais  polyg-lotta   (Vieillot).     The  Short- wing-ed  Yel- 
low Tree-Warbler. 

Spanish.  Almendrita  de  verano. 

"  Arrives  and  crosses  to  Europe  in  April,  returning  in  August 
and  September,  many  remaining  to  nest  around  Tangier." — 
Fav  ier. 

This  Yellow  Warbler  is  exceedingly  plentiful  near  Gibraltar, 
being  one  of  the  latest  of  the  spring  arrivals ;  the  first  observed 
was  on  the  25th  of  April,  and  the  earliest  date  on  which  eggs 
were  found  was  on  the  14th  of  May.  The  birds  frequent  trees 
and  bushes,  especially  willows  and  sallows ;  and  the  nest,  neatly 
built  and  cup-shaped,  in  a  great  measure  composed  of  sallow- 
cotton  and  thistle-down,  is  placed  in  bushes,  and  usually  contains 
four  pinkish-tinged  eggs,  marked  with  blackish  spots. 

This  bird  was  figured  by  Yarrell  as  the  Melodious  Willow- 
Warbler  (H.  icterina),  a  slightly  larger  species. 

First  primary  small,  but  longer  than  the  primary  coverts ;  in  H.  icterina 
it  is  scarcely  as  long,  but  the  wing  is  longer. 

Above  olive-green ;  below  lemon-yellow  ;  lores  yellow ;  secondaries 
broadly  margined  with  buffish  white.  The  inside  of  the  mouth  bright 
orange-yellow.  Length  4£  inches. 


47.    Hypolais    opaca    (Lichtenstein) .      The   Western    Pallid 
Warbler. 

"  This  Warbler  is  nearly  the  same  in  size  as  H.  polyglotta, 
but  somewhat  larger,  and  identical  with  that  bird  in  habits, 
times  of  arrival  and  departure,  and  also  in  manner  of  nesting. 
They  build  on  trees,  bushes,  and  small  plants,  laying  in  May  or 
the  beginning  of  June." — Favier. 

This  Pallid  Warbler  is  the  latest  of  all  the  spring  migrants 
that  arrive  in  Anclalucia,  being  a  little  later  than  H.  polyglotta. 
They  are  much  more  plentiful  eastward  of  Gibraltar  than  in  the 
immediate  vicinity,  where  they  are  rare.  Another  species, 

F 


66  TURDIDJS. 

H.  olivetomm,  is  stated  to  have  been  met  with  at  Tangier  and 
Fez,  but  did  not  come  under  my  observation,  and,  being  found  in 
S.E.  Europe  and  Asia  Minor,  is  not  likely  to  occur. 

Above   olive-brown  ;    tail  darker  brown  ;    throat  and  centre  of  belly 
white  ;  the  rest  below  buffy  white.     Length  5£  inches. 


Subfamily  REGULIN^. 

48.  Regulus-  cristatus,  K.  L.  Koch.    The  Gold-crested  "Wren. 

I  have  never  met  with  this  bird  in  Andalucia,  where  they 
have  been  recorded  as  common,  and  Lord  Lilford  informs  me 
that  he  never  obtained  an  Andalucian  specimen.  Possibly  they 
may  occur  irregularly,  like  the  Siskin — that  is  to  say,  not  in  all 
winters  consecutively.  Arevalo  says  they  have  been  found  near 
Malaga  and  Granada. 

Male.  General  colour  above  olive-green,  a  black  streak  on  each  side 
of  the  orange-coloured  crown. 

female.  Less  bright  than  male.     Length  3^  inches. 


49.  Regulus  ignicapillus   (0.  L.  Brehm).     The  Fire-crested 
Wren. 

This  species  is  resident  and  common  in  the  Cork-wood  and  in 
the  wooded  valleys  at  the  back  of  Algeciraz,  coming  as  near  to 
Gibraltar  as  the  Malaga  Gardens,  close  to  San  Roque.  They 
nest  rather  early,  the  young  being  able  to  fly  on  the  15th  of  May. 

Male.  Much  as  last,  but  whitish  streak  over  the  eye,  and  a  third  black 
streak  through  eye  to  the  olive-green  nape ;  crown  flame-coloured. 
Female.  Less  bright  than  male.     Length  4  inches. 


TIMELIID^E.  67 


Family  TIMELIID^].         Subfamily  BRACHYPODIIN^. 

50.  Pycnonotus  barbatus  (Desfontaines).     The  White-vented 
Bulbul. 

Moorish.  Bou  lag-lag. 

"  Is  very  abundant  and  resident  around  Tangier.  When  the 
oranges  are  ripe,  they  are  always  to  be  heard  and  seen  chattering 
and  fighting  in  the  gardens.  They  nest  in  May,  June,  and  July, 
laying  from  three  to  four  eggs,  which  are  very  thin-shelled  and 
tender,  of  a  greyish-white  colour,  marbled  or  spotted  with  reddish 
spots  of  two  or  three  shades  of  brown  and  purple.  The  nest  is 
built  in  the  branches  of  fruit-trees  (orange-,  apricot-,  pear-,  &c.), 
and  is  shaped  like  those  of  the  Woodchat  Shrike,  coarsely 
interlaced  outside  with  ends  of  small  roots  and  with  creeping 
plants.  They  feed  on  all  kinds  of  fruit  and  different  flowers, 
are  very  fond  of  oranges,  and  prefer  them  to  anything  else. 

"  This  species  is  subject  to  variations,  as  I  have  seen  two 
which  had  the  head,  breast,  and  neck  brown,  with  white  spots, 
while  the  wings,  back,  and  tail  were  brownish  red,  the  rest  being 
dirty  white." — Favier. 

In  accordance  with  Favier's  statement,  this  Dusky  Bulbul  was 
found  in  great  plenty  about  the  gardens  just  outside  Tangier. 
They  were  shy  ;  but  one  day  in  March  I  stalked  up  to  and 
watched  for  some  time  a  lot  of  seven  or  eight  in  the  Belgian 
Consul's  garden.  They  were  squabbling  and  playing  with  one 
another  on  a  Persian  lilac  or  common  bead- tree,  the  seeds  of 
which  they  were  pecking  at;  and  they  reminded  me  much  of 
some  of  the  Indian  Babblers  (Crateropus),  particularly  in  their 
flight  and  garrulous  chattering.  Besides  this  noise  they  have  a 
melodious  whistle,  which  I  took  down  at  the  time  and  tried  to 
note  thus — Pwit,  Pivit,  Quiterd,  Qiiiterd,  rather  in  the  tone  of  a 
Blackbird.  This  song,  if  it  may  be  so  called,  and  their  chatter 
are  so  remarkable  as  to  attract  attention  at  once. 

F2 


68  TIMELIID^E. 

I  took  a  good  deal  of  pains  to  ascertain  the  correct  local 
Arabic  name,  which  is  "  Bon  lag-lag  " ;  and  as  no  one  could  tell 
me  the  meaning  of  the  latter  part,  I  conclude  it  is  suggestive  of 
their  cry,  or  rather  clacking :  one  of  the  Arabic  names  of  the 
White  Stork  is  "  Bou  lak-lak,"  from  the  noise  made  by  the 
clacking  their  bills. 

Among  the  Jews,  who  speak  Spanish,  they  go  by  the  name  of 
"  Naranjero  "  (litt.  "  The  orange-man  "),  from  their  orange-eating 
propensities.  They  make  a  small  hole  in  the  side  of  an  orange 
and  completely  clean  it  out,  leaving  nothing  but  a  .shell  of 
orange-peel,  which  remains  hanging  on  the  tree.  I  have  more, 
than  once  pulled  these  husks  down,  thinking  them  to  be  sound 
fruit.  Owing  to  the  mischief  they  thus  do,  they  are  not 
favourites,  and  consequently  are  more  timid  near  Tangier  than 
about  Larache,  where  we  shot  some  of  them. 

We  were  informed  that  they  do  not  breed  till  the  end  of  May, 
and  so  had  no  opportunity  of  studying  their  nesting-habits.  In 
the  end  of  April,  near  Larache,  they  were  evidently  not  then 
nesting ;  and,  as  at  Tangier,  all  those  which  we  saw  were  near 
gardens  and  villages. 

Mr.  Meade- Waldo  found  them  in  February  1892,  out  in  the 
open,  feeding  on  the  flower  of  the  Asphodel. 

This  Bulbul  certainly  does  not  occur  in  the  western  part  of 
Andalucia  ;  I  have  tried  everywhere  for  it.  If  found  anywhere, 
the  coast  near  Tarifa  would  be  the  most  likely  ground ;  but  in 
the  orange-groves  there,  the  Spaniards,  when  asked  if  there  was  a 
bird  like  the  "  mirlo  "  which  ate  oranges,  simply  looked  on  me  as 
more  "loco"  than  the  generality  of  "los  Ingleses,"  who,  in  their 
opinion,  are  all  mad,  and  disclaimed  any  knowledge  of  a  "  naran- 
jero  "  in  the  shape  of  a  bird  of  such  size.  The  Great  Titmouse, 
however,  they  say  eats  oranges,  but  not  unless  the  skin  is  broken. 

Sexes  alike.  General  colour  dull  brown,  except  the  white  belly  and 
under  tail-coverts,  the  latter  sometimes  faintly  tinged  with  yellow.  Length 
8  inches,  tail  4  inches. 


CISTICOLA  CURSITANS.  69 


Subfamily  CISTICOL^E. 

51.  Cisticola  cursitans  (Frankland).      The  Fantail  or  Grass- 
Warbler. 

Moorish.  Bou-fesaha  (Father  of  eloquence).  Spanish.  Cierra- 
puiio,  Tin-Tin,  Biutrecillo,  Trepatorres,  Buitron. 

"  Is  the  most  common  of  the  aquatic  Warblers  around  Tangier, 
and  seen  migrating  in  lots  of  from  ten  to  twelve  during  March 
and  April,  returning  in  October,  November,  and  December. 
Many  remain  to  breed,  nesting  twice  in  the  season." — Favier. 

This  diminutive  Grass- Warbler  is  resident  near  Gibraltar,  and 
exceedingly  plentiful  in  the  winter,  frequenting  marshy  ground 
wherever  there  is  any  herbage,  such  as  grass,  sedges,  or  short 
rushes.  In  the  spring  they  go  to  the  corn-fields  as  well,  never, 
however,  being  found  away  from  water.  I  do  not  recollect  ever 
seeing  them  perch  on  a  bush  or  tree,  but  always  on  some  plant. 
Their  note  and  jerky  flight  somewhat  remind  one  of  the  Meadow- 
Pipit  ;  during  the  nesting-season  in  particular  they  will  fly 
darting  about  high  overhead  for  several  minutes,  continually 
uttering  their  squeaky  single  note  (whence  the  name  of  Tin-Tin), 
all  the  time  evidently  trying  to  decoy  the  intruder  from  their 
nest.  They  undoubtedly  breed  twice  a  year — according  to  the 
Spaniards,  three  times.  I  have  found  the  young  well  able  to  fly, 
and  a  nest  with  eggs  ready  to  hatch,  on  the  same  day,  the  19th 
of  April ;  an  unfinished  nest  on  the  8th  of  May,  and  a  nest  with 
eggs  very  hard  sat-on  on  the  10th  of  that  month. 

The  nest  much  resembles  the  caterpillars'  webs  which  are 
common  on  pine-trees  in  some  parts  of  Andalucia  ;  any  one  would 
take  them  for  the  web  of  an  insect ;  but  they  are  very  troublesome 
to  find,  and  made  of  the  cotton  of  plants  and  thistle-down, 
with  small  bits  of  grass  beautifully  sewn  and  interwoven  with 
the  corn  or  grass  in  which  the  nest  is  built ;  the  entrance  is  at 


rU  PANURID^E. 

the  top,  the  bottom  being  the  broadest  part,  the  whole  length 
about  five  inches.  The  usual  number  of  eggs  is  five,  generally 
of  a  pale  blue ;  but,  as  is  well  known,  they  vary  strangely  in 
colour. 

Sexes  alike.  Tail  much  graduated.  Inside  of  mouth  black  ;  irides  very 
pale  brown.  Crown  and  back  streaked  with  black  and  buff.  Below  buffisli 
white  ;  flanks  rufous.  Length  4^  inches,  tail  If  inches. 


CRATEROPODES. 

52.    Argya     fulva     (Desfontaines) .         The    North  -  African 
Babbler. 

This  Babbler  is  mentioned  by  Mr.  Drake  as  occurring  in  the 
southern  part  of  Morocco,  but  does  not  appear  ever  to  have  come 
under  Favier's  notice  in  the  northern  part. 

Sexes  alike.     Bill  stout  and  strongly  curved. 

General  colour  rufous  fawn.  Chin  and  throat  white.  Length  10 
inches,  tail  5  inches. 


PANURID^l. 

53.  Panurus  biarmicus  (Linnaeus).    The  Bearded  Reedling-. 

This  reed-frequenting  bird  is  found  at  the  Albufera  near 
Valencia,  but  I  have  no  certain  evidence  of  its  occurrence  nearer 
to  Gibraltar. 

Tail  fawn-coloured,  about  three  inches  long,  graduated,  and  slightly 
curving  downwards. 

Male.  Bill  yellow  ;  head  grey  ;  black  moustachial  stripe.  General 
colour  above  tawny  buff. 

Female.  Paler  ;  no  moustachial  stripe  ;  head  fawn-colour. 

Young.  Like  female,  but  head  and  back  slightly  striped  with  black. 
Length  6  inches. 


PARID.E.  71 


PARID.E. 

54.  Acredula  irbii,   Sharpe  and  Dresser.    The  Spanish  Long- 

tailed  Tit. 

Spanish.  Mito. 

This  little  bird,  a  climatic  race  of  our  British  A.  caudata  (or,  as 
separatists  have  it,  rosea),  is  only  to  be  found  around  Gibraltar  in 
the  Cork-wood  of  Almoraima,  chiefly  keeping  to  the  sotos  and  to 
the  district  round  the  Mill,  the  Long  Stables,  and  the  second 
venta.  Similar  in  its  habits  to  the  British  form,  the  nest  and 
eggs  are  also  exactly  the  same  as  those  of  that  bird.  I  found  the 
young  able  to  fly  by  the  middle  of  April,  and  on  the  12th  of  that 
month  found  a  nest  with  seven  young  fully  fledged ;  this  would 
make  the  date  of  laying  about  the  20th  of  February.  The  nests, 
without  exception,  were  all  built  in  the  thorny  creeper,  a  species 
of  Smilax,  called  Zarzaparilla  by  the  Spaniards,  which  forms 
regular  net-  or  lattice-work  walls  from  the  ground  to  the  lower 
branches  of  the  trees,  and  are  usually  placed  about  15-16  feet 
from  the  ground  and  very  difficult  to  get  at,  the  only  way  being 
either  to  cut  or  shoot  away  the  creepers  above  them — often  no 
easy  matter.  The  only  eggs  obtained  were  addled  ones,  left  in 
nests  from  which  the  young  had  flown. 

The  adults  differ  from  the  British  and  North-European  races  in  having 
the  entire  back  bluish  grey. 

Crown  black,  with  whitish  stripe  along  the  centre.  Length  5^  inches, 
tail  3|  inches. 

55.  Parus  major,  Linnseus.    The  Great  Tit. 

Spanish.  Quive-vive,  Carpintero,  Carbonero,  Cerrajero,  Guer- 
rero. 

Favier  considers  this  species  to  be  extremely  scarce  near 
Tangier,  but  specimens  have  been  obtained  there  in  winter,  and 
Mr.  Meade-Waldo  saw  plenty  on  the  27th  of  February,  1892. 


72  PAEID^E. 

On  the  Spanish  side  of  the  Straits  they  are  extremely  plentiful, 
and  to  be  heard  wherever  there  are  any  trees,  nesting  in  April 
in  holes  of  the  cork-trees.  There  is  a  great  increase  of  their 
numbers  in  winter,  when  they  visit  the  Alameda  and  gardens  at 
Gibraltar,  being  the  only  observed  Calpeian  representative  of  the 
Tits. 

Male.  Crown  bluish  black  ;  cheeks  white  ;  black  stripe  along  the  middle 
of  the  greenish-yellow  breast  and  belly. 

Female.  Similar,  but  black  stripe  below  not  so  extended.  Length  6 
inches. 

56.  Parus  cseruleus,  Linnseus.    The  Blue  Tit. 

Spanish.  Herrerillo  (Little  smith). 

The  Blue  Tit  is  very  common  in  Andalucia,  being  resident  and 
particularly  abundant  in  the  Cork-wood  of  Almoraima,  generally 
nesting  (about  the  middle  of  April)  in  the  decayed  hollow 
branches  of  the  cork-trees. 

In  April  1894,  at  Retin,  I  saw  a  pair  of  these  birds  going  in 
and  out  of  a  dome-shaped  nest,  constructed  externally  of  small 
sticks,  which  they  had  either  built  or  enlarged,  at  the  end  of  a 
horizontal  bough  of  a  cork-tree.  It  looked  like  a  miniature 
Magpie's  nest,  and  I  could,  when  it  was  first  found,  see  the  light 
through :  unable  to  reach  it  without  a  ladder,  I  left  for  Gibraltar, 
and  Verner,  who  remained  behind  for  a  few  days,  reported 
that  ultimately  they  began  to  dismantle  the  nest  and  carry  the 
materials  to  some  more  usual  situation. 

Spanish  specimens  are  very  bright  in  colour — one  or  two  so  much  so 
that,  until  I  had  seen  a  specimen  of  Parus  tenerffce,  I  imagined  them  to 
be  that  species. 

Crown  and  wing-coverts  azure-blue.     Length  4^  inches. 

57.  Parus  teneriflfee,  Less.    The  Ultramarine  Tit. 

Moorish.  Bou  rezizi  (Father  of  the  little  turban). 

"  Is  resident  near  Tangier,  but  less  frequent  in  December  and 


PAKUS  CRISTATUS.  73 

January  than  during  other  months.  They  nest  in  holes  of  trees, 
in  April,  laying  from  four  to  five  eggs,  white,  with  very  small  red 
spots,  similar  to  those  of  Parus  cceruleus,  which  they  replace  in 
Morocco,  and  appear  to  be  a  variety  of  that  bird  constant  to  this 
climate." — Favier. 

I  have  never  been  able  to  detect  the  Ultramarine  Tit  on  the 
Spanish,  nor  seen  the  Blue  Tit  on  the  Moorish  side  of  the  Straits, 
where  the  present  species  is  plentiful,  their  habits  &c.  being 
identical. 

The  African  bird  is  easily  recognized  by  the  greyish-blue  back  and  the 
deep  blackish  blue  on  the  crown  of  the  head,  as  well  as  on  those  parts  which 
are  cobalt-blue  in  the  European  species. 

58.  Parus  ater,  Linnaeus.    Trie  European  Coal-Tit. 

I  never  met  with  this  species,  which,  however,  is  recorded 
from  Granada,  Cordova,  and  Malaga  by  Arevalo. 

The  Algerian  Coal-Tit  (Parus  ledouci]  in  all  probability  is  to 
be  met  with  in  Morocco.  This  bird  has  those  parts  of  the  head 
and  nape  of  the  neck  lemon-yellow  which  are  white  in  P.  ater, 
the  under  surface  being  also  lemon-yellow. 

Head  black  ;  nape  and  cheeks  white  ;  wing-coverts  tipped  with  white, 
forming  two  bars  on  the  wing  ;  back  bluish  gre.y  in  winter.  Length  4£ 
inches. 

59.  Parus  palustris,  Linnaeus.    The  Marsh-Tit. 

I  did  not  observe  this  bird  near  Gibraltar,  but  Mr.  Saunders, 
Mr.  Stark,  and  Arevalo  all  report  it  from  near  Granada. 

Back  greyish  brown  ;  crown,  chin,  throat,  and  nape  glossy  black. 
Length  4^  inches. 

60.  Parus  cristatus,  Linnaeus.    The  Crested  Tit. 

Spanish.  Capuchino. 

The  Crested  Tit  is  resident  and  common  in  the  Cork-wood  of 
Almoraima,  in  all  the  neighbouring  pine-woods,  and  in  the 
valleys  and  on  the  hill-sides  at  the  back  of  Algeciraz  up  to  near 


74  SITTID^E. 

Tarifa,  wherever  the  cork-tree  grows  ;  but  singularly  is  not  found 
in  the  Sierra  Retin,  or  in  the  Sierra  Bartolome.  They  nest  about 
the  10th  of  May,  in  the  hollow  ends  of  boughs  of  the  cork-  and 
pine-trees,  the  eggs  being  about  five  in  number,  much  spotted, 
and  resembling  strongly  those  of  the  Creepei(Certhiafamilians). 
There  is  reason  to  think  the  species  occurs  in  Morocco,  but 
this  is  only  mentioned  with  the  view  of  directing  the  attention  of 
future  collectors  there,  in  order  that  they  may  look  out  for  them. 

General  colour  greyish  brown  ;  throat  black  ;  crown  dull  black,  all  the 
feathers  tipped  and  edged  with  white,  and  the  hinder  feathers  long  and 
pointed,  forming  a  conspicuous  crest.  Length  4£  inches. 

61.  -iEglthalus  pendulirms  (Linnseus).    The  Penduline  Tit. 

Spanish.  Pajaro  moscon. 

Specimens  of  this  bird  have  been  procured  near  Malaga ; 
further  east  they  are  more  common,  building  their  retort-shaped 
nest  at  the  extremity  of  boughs,  generally  of  poplars. 

Bill  short,  straight,  and  much  pointed. 

Crown  and  nape  greyish  white  ;  forehead  and  sides  of  head  black. 
Above   chestnut-brown  ;    throat   white  ;    rest   of    underparts   creamy. 
Length  4^  inches. 


SITTIDJE. 

62.  Sitta  csesia,  Wolf.    The  Common  Nuthatch. 

This  bird,  our  British  Nuthatch,  was  not  mentioned  by  Favier, 
but  Capt.  S.  Reid  mentions  five  or  six  having  been  obtained  by 
Olcese  near  Tangier  about  1883. 

On  the  Spanish  side,  Mr.  Saunders  records  them  to  be  common 
near  Granada,  as  also  does  Arevalo,  stating  them  to  be  resident. 
We  did  not  meet  with  any  near  Gibraltar. 

Bill  longer  than  head.  Above  slaty  grey ;  throat  whitish  ;  rest  of 
underparts  cinnamon-buff,  lightest  next  the  throat  ;  legs  dull  brown. 
Length  5|  inches. 


TROGLODYTID.E.  75 

CERTHIIDJE. 

63.  Certhia  familiaris,  Linnaeus.    The  Tree-Creeper. 

Spanish.  Barba-jelena,  Trepa-troncas,  Aranero. 

The  Tree-Creeper  is  resident  and  common  in  the  Cork-wood 
and  in  the  valleys  near  Algeciraz,  nesting  in  April. 

A  single  specimen  was  shot  near  Tangier  about  the  20th  of 
April,  the  only  one  I  heard  of  on  the  African  side. 

Bill  slender,  curved  downwards,  and  pointed.  Above  brown,  spotted 
with  paler  brown  ;  below  white.  Tail  graduated  or  cuneiform,  reddish 
brown,  with  stiff  points.  Length  5  inches. 

64.  Tichodroma  muraria,  Linnaeus.    The  Rock-Creeper. 

I  have  never  met  with  this  bird,  which  is  recorded  from  near 
Antiquera,  Gaitan,  and  the  Sierra  Nevada,  and  is  stated  to  have 
been  seen  at  Gibraltar  many  years  ago. 

General  colour  slate-grey  ;  all  the  primaries,  except  the  first  three, 
crimson  on  the  basal  half  of  the  outer  web  ;  throat  black  in  summer,  white 
in  winter  ;  bill  as  in  last,  but  longer  in  proportion.  Length  6  inches. 

TROGLODYTID.E. 

65.  Troglodytes  parvulus,  K.  L.  Koch.    The  Wren. 

Spanish.  Cucito,  Katilla. 

"  Resident  near  Tangier,  and  numerous,  nesting  from  March 
to  June.  Some  are  migratory,  arriving  in  November,  and  leaving 
again  in  February." — Favier. 

The  above  remarks  equally  apply  to  the  Wren  on  the  Spanish 
side,  where  they  are  most  abundant,  nesting  very  early ;  and  I 
have  seen  young  fully  able  to  fly  on  the  26th  of  April,  and  they 
are  resident  on  "  the  Rock." 

It  is  very  curious  that  this  little  bird  should  be  a  resident  in 


76  ORIOLID^E. 

the  scorching  sun  of  Morocco  and  Andalucia  as  well  as  in  the 
bitter  cold  of  the  hills  in  Inverness-shire,  where  they  are  one  of 
the  very  few  birds  which  remain  to  brave  the  winter. 

Above  reddish  brown  ;  breast  whitish  brown  ;  tail  closely  barred  with 
black,  and  much  more  rufous  than  back.  Hind  toe  as  long  or  longer  than 
middle  toe.  Length  4  inches. 


CINCLID.E. 

66.  Cinclus  albicollis  (Vieillot).     The   Grey-backed   Dipper 
or  "Water-Ouzel. 

Spanish.  Tordo  de  agua,  Pechiblanco. 

Occurs  in  the  streams  of  the  Sierras,  and  is  resident.  I  have 
seen  them  near  the  waterfall  beyond  Algeciraz,  where  Mr.  Stark 
found  a  nest  about  the  17th  of  May ;  but  they  are  not  abundant 
anywhere. 

Resembles  the  English  Chestnut- breasted  Dipper,  C.  aquaticus,  except 
that  the  back  is  paler  and  the  head  lighter,  and  it  can  only  be  considered  a 
climatic  race  of  that  species.  Length  7  inches. 


ORIOLIDJE. 

67.  Oriolus  galbula,  Linnaeus.    The  Golden  Oriole. 

Moorish.  Tair  es  sfar  (Yellow  bird).     Spanish.  Oropendola. 

According  to  Favier  the  Golden  Oriole  "  crosses  the  Straits  in 
great  numbers  during  April  and  May,  returning  in  July,  August, 
and  September."  These  dates  much  agree  with  my  own  obser- 
vations on  the  Spanish  side,  I  having  first  seen  them  in  1869  on 
the  21st  of  April,  in  1870  on  the  18th,  in  1871  on  the  4th  (one 
only),  and  in  1872  on  the  llth  :  many  passed  on  the  16th  in  that 
year.  The  spring  migration  lasts  up  to  the  14th  or  15th  of  May. 
Some  few  pairs  remain  to  breed  in  the  vicinity  of  Gibraltar ;  but 
the  majority  pass  further  north  and  resort  to  fruit-producing 
districts,  where  they  get  the  credit  of  doing  much  damage  to 


STURNID^E.  77 

cherries,  mulberries,  &c.,  but  at  the  same  time  they  eat  cater- 
pillars and  other  insects. 

Almost  entirely  a  fruit-eating  bird ;  those  who  have  kept  them 
alive  informed  me  that  they  could  not  preserve  them  through  the 
winter — nor,  indeed,  longer  than  fruit  was  to  be  obtained. 

When  the  Japanese  loquats  (Eriolotrya  japonica)  were  ripe  in 
my  garden  at  Gibraltar  in  May  1870,  Golden  Orioles  remained 
about  as  long  as  the  loquats  lasted,  but  would  not  admit  of  much 
observation,  as  they  were  very  shy  and  difficult  to  watch.  They 
are  more  often  heard  than  seen ;  and  I  have  spent  hours  in 
trying  to  get  a  shot  as  they  skulked  in  the  thickest  foliage  of  tall 
trees,  continually  piping  their  flute-like  note. 

Some  are  always  to  be  heard  during  May  near  the  Mill  and  the 
"  Second  Veuta  "  in  the  Cork-wood  ;  and  a  pair  usually  frequented 
the  lower  part  of  the  First  Pine-wood.  I  found  one  nest  in  the 
middle  of  May,  built  at  the  very  extremity  of  a  bough  at  the  top 
of  a  high  oak  tree,  which  was  impossible  to  obtain  without 
cutting  the  branch  off.  Verner  remarks  also  that  the  Golden 
Oriole  has  a  habit  of,  when  disturbed,  flying  into  the  thickest 
green-foliaged  tree,  not  settling  on  cork-trees,  in  which  the 
foliage  is  scanty. 

Male.  Head  and  general  colour  golden  yellow.  Black  spot  between 
crimson  eye  and  dull  red  bill  ;  wings  black,  with  yellow  tips  to  secondaries. 
Tail  :  outer  feathers  with  more  than  terminal  half  yellow. 

Female.  Greenish  yellow  above  ;  outer  tail-feathers  tipped  with  yellow. 
1  have  never  seen  a  female  Golden  Oriole  in  same  plumage  as  male. 

Young.  Greenish  yellow  above  ;  below  whitish  tinged  with  yellow  ; 
breast  slightly  streaked  with  brown  ;  the  black  parts  of  adult  male  replaced 
by  greenish  brown.  Length  9  inches. 

STURNID53. 

68.  Sturnus  vulg-aris,  Linnaeus.    The  Common  Starling. 

Moorish.  Zarzor.     Spanish.  Estornino. 

"  This  bird  arrives  about  Tangier  in  large  flights  from  October 


78  STURNID^E. 

to  January,  departing  in  March.  During  the  autumnal  migration 
the  flights  are  often  mixed  up  with  S.  unicolor.  In  October, 
1842,  a  Moor  brought  to  Tangier  about  three  hundred  and  fifty 
Starlings,  which  he  affirmed  he  had  caught  at  one  time  in  a  net ; 
about  half  of  these  birds  were  S.  unicolor." — Favier. 

In  Andalucia,  the  Common  Starling  may  be  said,  roughly 
speaking,  to  come  and  go  with  the  Golden  Plover.  The  earliest 
date  noticed  of  their  arrival  was  the  15th  of  October,  the  latest 
date  on  which  any  were  seen  being  the  1st  of  March.  I  have  a 
note  of  seeing  many  thousands  passing  southwards  in  successive 
flights  on  the  28th  of  October.  During  the  winter  months  they 
are  seen  in  swarms  about  low  ground ;  and  the  Spaniards  shoot 
immense  numbers  at  their  roosting-places  in  the  reed-beds  near 
Vejer  and  Casas  Viejas.  Consequently,  during  their  stay, 
Starlings  form  a  very  cheap  and,  it  may  be  fairly  said,  nasty  dish 
in  all  the  ventorillas  in  the  vicinity. 

Male.  Black,  with  purple  and  green  gloss,  small  buff  tips  to  feathers 
above.  Head,  throat,  and  breast  with  green  gloss. 

Female.  Similar  to  male,  but  less  brightly  marked. 

In  ivinter,  all  the  feathers  of  upper  parts  tipped  with  buff,  those  below 
tipped  with  whitish. 

Young  in  first  plumage.  Uniform  greyish  brown  ;  throat  dull  white  ; 
lower  parts  clouded  with  white.  Length  8^  inches. 


69.  Sturnus  unicolor,  Temminck.    The  Spotless  Starling". 

Moorish.  Zarzor  kahal  (Black  Starling).     Spanish.  Tordo. 

"  This  Starling  is  very  abundant  around  Tangier,  passing  north 
in  March  and  returning  during  the  month  of  September,  many, 
however,  remaining  to  breed." — Favier. 

The  Sardinian  Starling,  as  the  species  has  been  termed, 
Spotless  Starling  being  a  more  appropriate  and  distinctive  name, 
is  almost  entirely  migratory  in  Andalucia  ;  but  I  have  seen  them 
there  in  December.  Not  so  abundant  as  the  Common  Starling, 
they  resemble  that  bird  in  their  habits  and  note,  nesting  about 


PASTOE  KOSEUS.  79 

the  end  of  April  in  roofs  of  houses  in  towns,  and  they  make  use 
of  the  old  Moorish  towers,  besides  building  in  holes  of  trees  ; 
the  eggs  exactly  resemble  those  of  8.  vulgar  is.  They  are  more 
common  some  sixty  miles  north  of  Gibraltar  than  in  the  imme 
diate  vicinity.  'Three  or  four  pairs  used  to  frequent  the  Venta 
at  Casas  Viejas,  and  during  November  and  December  nearly 
every  morning  assembled  on  the  roof,  whistling  and  pluming 
themselves  before  going  forth  for  the  day.  The  amo,  or  landlord, 
well  known  as  "  old  Bernardo,"  begged  me  not  to  molest  them — 
a  request  I  most  scrupulously  complied  with  ;  but  on  my  return 
there  in  1874,  they  were  absent,  probably  killed  by  some  of  the 
shooting  visitors  from  Gibraltar.  Whether  the  death  of  the  old 
man  caused  them  to  lack  protection  I  cannot  say.  This  old 
fellow,  who  had  served  as  a  sergeant  in  the  Spanish  army,  and 
was  present  at  the  defence  of  Tarifa  in  1811,  was  a  fine  specimen 
of  the  Spaniard,  and  used  to  tell  wonderful  stories  of  his 
soldiering  days.  I  regret  that  since  his  decease  the  Venta  has 
changed  for  the  worse,  both  in  prices  and  accommodation. 

Head  and  back  uniform.  No  spots  in  breeding-plumage,  but  in  winter 
all  the  body-feathers  are  tipped  with  arrow-shaped  spots  of  greyish  white. 
Length  8^  inches. 

70.  Pastor  roseus  (Linnaeus) .    The  Rose-coloured  Starling. 

The  Rose-coloured  Pastor,  as  it  used  to  be  termed,  is  an 
occasional  wanderer  from  the  East  to  Andalucia,  and  has  been 
obtained  near  Seville  and  Malaga. 

Male.  Head  black,  with  crest  of  pointed  and  elongated  feathers  ;  back, 
scapulars,  rump,  breast,  sides,  and  belly  rosy  pink;  wings,  tail,  and  thiglis 
black  ;  bill  rose-coloured,  black  at  base. 

Female.  Crest  smaller  and  less  bright. 

Young.  Crest  absent  ;  colour  greyish  brown,  where  rosy  in  adults. 
Length  8^  inches. 


80  CORVID^E. 


GORVID^l. 

71.  Pyrrhocorax     graculus     (Linnaeus).        The     Red-billed 

Chough. 

Moorish.  Narrar.     Spanish.  Graja. 

This  species  is  stated  by  Favier  to  be  found  in  large  nights 
near  Tetuan — a  statement  we  can  fully  corroborate.  We  also 
saw  a  great  many  about  the  cliffs  of  Abyla,  or  Apes'  Hill,  opposite 
to  Gibraltar. 

In  the  rocky  sierras  of  Andalucia  the  Red-billed  Chough  is 
plentiful,  particularly  about  Ubrique.  They  are,  of  course, 
resident ;  but  I  am  unable  to  state  the  time  of  nidification. 

Jet-black,  with  steel  gloss  ;  the  curved  bill,  eyelids,  legs,  and  feet  red. 
Length  14-17  inches. 

72.  Pyrrhocorax     alpinus,     Vieillot.       The     Yellow  -  billed 

Chough. 

Although  there  is  no  record  of  the  occurrence  of  this  Chough 
on  the  African  side,  at  Apes'  Hill,  in  1877,  a  bird  flew  so  close 
to  us  that  we  thought  we  could  distinguish  the  yellow  bill. 

On  the  Spanish  side  they  occur  near  Granada,  and  at  the 
Taj os  del  Gaitan,  Malaga. 

Plumage  and  size  as  last,  but  bill  yellow  and  slightly  shorter. 

[Nucifrag'a  oaryooatactes  (Linnaeus).    The  Nutcracker. 

I  never  met  with  this  pine-forest  bird  on  either  side  of  the 
Straits,  though  there  was  one  in  a  collection  at  Cordova ;  but 
from  what  locality  was  not  stated ;  so,  until  further  evidence  be 
obtained,  the  species  cannot  with  certainty  be  included  as  an 
Andalucian  bird.  It  has  been  recorded  from  Estremadura  by 
Captain  Cook  Widdrington,  and  possibly  may  be  found  in  some  of 
the  high  ranges,  where  Pinus  cembra  grows.] 


GAKRULUS  GLAN DARIUS.  81 

Head  dark  brown  ;  general  colour  clove-brown,  most  of  the  feathers  of 
breast,  sides,  and  underparts  spotted  with  a  triangular  white  mark.  Rump 
and  upper  tail-coverts  uniform  brown.  Tail  blackish,  tipped  with  white. 
Length  14  inches. 


73.   Garrulus  glandarius    (Linnseus).      The    "West-European 
Jay. 

Spanish.  Arrendajo. 

This  British  Jay  is  not  record3d  by  Favier  from  Morocco ;  nor 
did  I  obtain  any  species  of  Jay  on  the  African  side ;  but  Captain 
Savile  Reid,  in  February  1883,  brought  home  a  specimen,  one  of 
three  obtained  by  Olcese  near  Tangier.  The  Editors  of  the 
'  Ibis,'  1885,  p.  246,  examined  this  bird,  and  considered  it  a  cross 
between  G.  glandarius  and  G.  cervicalis.  We  may  reasonably 
assume  that  they  sometimes  cross  the  Straits,  as  they  occasionally 
appear  at  Gibraltar  in  winter.  Four  frequented  the  Alameda 
and  other  gardens  in  the  south  from  about  the  10th  of  November, 
1870,  to  the  4th  of  April,  1871 ;  and  I  saw  another  in  March, 
1872 ;  this  last  bird  did  not  linger  about  for  more  than  a  few 
days. 

This  Jay  is  very  plentiful  near  Gibraltar  in  the  Cork-wood,  and 
in  the  wooded  valleys  and  hill-sides  up  to  a  considerable  elevation. 
At  the  same  time  it  is  rather  local ;  and  though  many  are 
resident,  they  are  more  abundant  in  the  winter  months. 

They  nest  in  some  numbers  in  the  Cork-wood,  laying  their 
eggs  early  in  May ;  and,  at  that  season  particularly,  they  are 
easily  decoyed  within  shot  by  secreting  one's  self  in  thick  cover 
and  imitating  either  their  call  or  the  squeal  of  a  wounded  rabbit. 

Throat  white  ;  crown  with  black  streaks  ;  feathers  of  forehead  and  crown 
edged  with  white  ;  back  vinaceous  ;  outer  web  of  wing-coverts  barred  with 
blue  and  Hack,  inner  web  black  ;  tail  often  barred  slightly  at  the  basal  end 
in  the  same  manner  as  the  outer  web  of  wing-coverts  (this  is  not  the  result 
of  age,  birds  in  first  plumage  occasionally  being  so  marked)  ;  iris  pale  blue. 
Length  14  inches. 


82  COKVID^E. 

74.  Garrulus  cervicalis,  Bonaparte.    Algerian  Black-headed 

Jay. 

Mr.  Meade-Waldo  found  this  bird  on  the  Moorish  side  of  the 
Straits,  and  brought  home  several ;  he  describes  it  as  chiefly  a 
mountain  bird,  with  a  voice  different  to  that  of  G.  glandarius. 

Crown  with  thick  black  crest ;  nape  and  fore  part  of  back  vinous  rufous  ; 
space  round  eye  white.  Length  14  inches. 

75.  Pica  rustica  (Scopoli).    The  Common  Magpie. 
Spanish.  Urraca,  Marica. 

Our  British  Magpie  is  extremely  local  in  Andalucia ;  but 
where  met  with  is  very  abundant ;  they  do  not,  however,  occur 
to  the  south  of  Seville,  except  on  the  banks  of  the  Guadalquivir 
to  below  Coria,  as  far  as  there  are  any  trees  and  bushes.  Great 
numbers  frequent  the  Goto  del  Hey,  where  they  breed  in  the 
beginning  of  May,  accommodating  the  Great  Spotted  Cuckoo 
with  their  nests.  The  Spanish  bird  undoubtedly  runs  into  the 
African  form  P.  mauritanica. 

Long  graduated  iridescent  green  tail ;  rump  greyish  white.  Length 
16-18  inches,  tail  10-11. 

76.  Pica  mauritanica,  Malh.    The  North- African  Magpie. 

Arabic.  Akaka. 

This  species,  which,  however,  I  failed  to  meet  with  in  Morocco, 
is  the  Magpie  of  the  country,  and  perhaps  is  very  local,  for 
Mr.  Drake  describes  it  as  abundant  in  parts  he  visited. 

Is  distinguished  from  P.  rustica  by  the  bare  space  behind  the  eye  and 
by  the  black  rump,  both  species  being  otherwise  identical  in  size  and 
markings. 


o' 


77.  Cyanopica  cooki,  Bonaparte.    The  Spanish  Azure-winged 
Magpie. 

Spanish.  Mohino  rabilargo,  Mohino. 

This  species  is  peculiar  to  the  Peninsula,  but  does  not  occur 
in  the  vicinity  of  Gibraltar.     The  nearest  locality  to  that  place 


>7        cfl 
£*. 


W     0 
D 
J 
CD 


7 


CORVUS  MONEDULA.  83 

where  they  are  to  be  found  is  about  Coria  del  Rio,  below  Seville  ; 
thence,  as  far  as  the  Goto  del  Rey,  they  occur  in  tolerable 
numbers,  but  are  much  more  common  towards  Cordova,  and  are 
reported  to  be  very  numerous  in  some  parts  of  Estremadura, 
being,  however,  a  very  local  bird. 

The  nests  which  I  have  seen  were  built  on  boughs  at  no  great 
height  from  the  ground,  rather  clumsily  constructed  with  small 
sticks,  grass,  moss,  and  wool — containing  five  eggs ;  but  as  many 
as  seven  are  frequently  found.  They  are  well  figured  in  '  The 
Ibis'  (1866,  p.  382,  pi.  x.  figs.  3-8),  from  specimens  obtained  by 
Lord  Lilford  in  1864.  and  vary  a  great  deal  in  colour  and 
markings,  the  commonest  form  being  of  a  stony-buff  colon?* 
marked  with  purplish  and  brown  spots.  I  kept  four  of  these 
birds,  reared  from  the  nest,  for  some  time  alive,  feeding  them  on 
grapes,  figs,  bread,  beetles,  and  grasshoppers.  Always  placing 
the  insects  under  their  feet,  they  picked  them  to  pieces  much  as 
a  Hawk  or  an  Owl  tear  their  prey.  They  became  very  tame  and 
amusing ;  but  during  my  temporary  absence,  unluckily,  all  died. 

I  never  heard  of  this  bird  on  the  Moorish  side  of  the  Straits. 

The  sexes  are  alike  in  plumage. 

78.  Corvus  monedula,  Linnaeus.    The  Jackdaw. 

Spanish.  Graja. 

Mr.  Drake  mentions  having  met  with  the  Jackdaw  near  Tetuan, 
where  I  did  not  see  any,  nor  find  them  anywhere  on  the  African 
side.  Favier  also  omits  the  bird  from  his  list. 

On  the  Spanish  side  of  the  Straits  they  are  extremely  local, 
the  only  locality  in  which  I  have  seen  any  being  the  Goto  del 
Rey,  near  Seville,  where,  in  1870,  they  were  common,  nesting 
about  the  end  of  April  in  holes  of  trees,  one  or  two  pairs 
building  in  the  roof  of  the  Palacio.  Jackdaws  are  also  said  to 
be  abundant  near  Granada. 

Black  ;  ear-coverts,  nape,  and  sides  of  neck  grey  ;  belly  leaden  black. 
Youny.  Little  or  no  grey  on  nape.     Length  14  inches. 

G2 


84  CORV1D.E. 

79.  Corvus  corax,  Linnaeus.    The  Raven. 

Spanish.  Cuervo. 

The  Raven  is  found  sparingly  but  very  generally  distributed  on 
the  Spanish  side  of  the  Straits,  but  does  not  seem  to  be  found  on 
the  Moorish  side.  They  are  resident,  and  usually  commence  to 
lay  about  the  middle  of  March,  thus,  very  curiously,  breeding 
later  in  Andalucia  than  in  England  or  further  north.  Verner 
found,  on  the  24th  of  May,  1878,  a  nest  with  young  just  hatched, 
a  few  days  earlier  one  with  fledged  birds,  whilst  in  1886,  at  Crete, 
he  took  fresh  eggs  of  Ravens  on  26th  of  March,  when  other 
nests  had  eggs  or  were  ready  for  eggs. 

One  pair  nest  at  Gibraltar,  and,  as  customary  with  Ravens,  are 
the  terror  of  all  birds  that  approach  their  domain.  Another  pair 
nest  at  Casas  Viejas,  in  the  old  quarry  called  La  Cima,  just 
outside  the  village. 

In  no  case  that  I  have  seen  have  their  nests  been  in  anything 
like  proximity  to  one  another,  the  reverse  being  the  case  with 
Corvus  tingitanus. 

General  colour  glossy  steel-black  ;  throat-feathers  pointed  ;  tail  slightly 
cuneate.  Length  24  inches. 

80.  Corvus  tingitanus,  Irby.    The  Tangier  Raven. 

Moorish.  Grab. 

This  species  or  race  appears  to  me  to  be  quite  distinct  from 
C.  corax,  and  was  noticed  and  described  in  'The  Ibis,'  1874 
(p.  264). 

Smaller  than  the  Common  Raven,  C.  corax,  its  note  is  different, 
while  its  very  gregarious  habits  are  opposed  to  those  of  our 
common  Raven ;  and  it  is  the  Raven  of  the  Canary  Islands. 

Many  specimens  are  very  much  marked  with  rusty  brown  on 
the  wings  and  tail,  others  very  slightly  so.  In  all  that  I  have  seen 
there  is  a  tinge  of  brown  on  the  wings — not  that  this  coloration 
is  of  any  consequence  in  determining  it  as  a  distinct  species. 
They  also  breed  later  than  C.  corax  does  on  the  Spanish  side. 


COBVUS  TINGITANUS.  85 

This  Raven  is  exceedingly  abundant  around  Tangier  and  along 
the  coast  as  far  as  some  distance  south  of  Larache.  We  did  not 
observe  any  in  the  high  parts  about  Apes'  Hill.  Outside  Tangier, 
flocks  of  them  may  be  seen  feeding  on  the  refuse  which  is  carried 
from  the  town  and  thrown  on  the  sea-shore.  They  are  exceedingly 
tame  to  the  natives,  being  viewed  with  superstitious  awe  by  the 
Moors,  but  are  wide  awake  to  the  European,  especially  if  he 
carries  a  gun,  and  if  once  fired  at  are  not  likely  to  give  a  second 
chance.  The  only  way  to  be  sure  of  getting  them  is  at  their 
nest,  which,  constructed  of  sticks,  neatly  lined  with  grass  and 
small  roots,  is  built  in  clefts  of  rocks,  on  trees  and  in  low  bushes ; 
one  nest  which  I  saw  was  fixed  in  the  crook  or  angle  formed  by 
a  dead  flowering  stalk  of  the  aloe  (Agave  americana),  which  had 
fallen  across  another  stalk  in  full  flower. 

The  eggs  are  usually  laid  about  the  20th  of  April,  and  vary  in 
number  from  five  to  seven,  and,  like  those  of  others  of  the  Crow 
tribe,  differ  much  in  the  markings.  Favier  in  his  MS.  says  of 
this  Raven,  under  the  head  of  C.  corax,  "  This  species  is  another 
of  those  birds  for  which  the  Mahometans  evince  a  superstitious 
feeling,  the  liver,  tongue,  brain,  and  heart  of  the  Raven  being 
considered  antidotes  against  the  effects  of  the  evil  one ;  the  same 
virtues  are  attributed  to  the  feathers  and  heart  of  the  Hoopoe. 
The  Raven  is  the  only  species  of  Crow  found  in  the  neighbourhood 
of  Tangier,  and  is  very  abundant." 

It  seems  that  this  African  Raven  crosses  the  Straits,  as  Verner 
reports  having,  near  Gibraltar,  taken  a  nest,  apparently  of  this 
bird,  but  was  unable  to  trap  or  shoot  an  old  bird.  He  also  reports 
having  seen  "  small  Ravens,"  and  "  forty  small  Ravens  were  seen 
together  near  Tarifa,"  "  fifty  Ravens  were  seen  together  "  on  the 
29th  of  April,  1880,  near  the  1st  River;  while  another  officer 
noticed  forty-two  together  near  Palmones  on  23rd  of  April. 

Resembles  the  Common  Raven,  but  has  the  throat-hackles  not  so  long, 
and  is  much  smaller  in  size.  Length  18*5  inches,  wing  14*5,  tail  8,  tarsus 
2*5,  bill  from  gape  2'5. 


86  CORVID.E. 

81.  Corvus  corone,  Linnaeus.    The  Carrion-Crow. 

Spanish.  Gragilla. 

Not  mentioned  by  Favier,  but  is  included  by  Mr.  Drake  in  his 
list  of  the  birds  of  Morocco  ;  I  never  met  with  any  on  the  African 
side.  On  the  Spanish  side  the  Black  Crow  is  scarce,  and  I  only 
remember  one  nest,  which  was  taken  near  Utrera  during  the 
month  of  March  ;  another  pair  nested  near  Motril  in  1893  (Mena). 

Black,  with  metallic  gloss  ;  nostrils  covered  with  bristly  feathers  ;  base 
of  body-feathers  whitish.  Length  18-19  inches. 

82.  Corvus  cornix,  Linnaeus.    The  Hooded  or  Grey  Crow. 

Not  recorded  from  the  Moorish  side  of  the  Straits.  Saunders 
mentions  the  Hooded  Crow  as  having  been  met  with  in 
Andalucia ;  there  was  one  in  a  museum  at  Seville,  but  upon 
inquiry  it  proved  to  be  from  France.  I  never  observed  any,  and 
the  Grey  Crow  must  be  a  very  rare  straggler,  as  Arevalo  does  not 
mention  it  as  Andalucian. 

Head,  throat,  wings,  thighs,  and  tail-feathers  black  ;  the  rest  of  plumage 
dull  grey.  Length  18-19  inches. 

83.  Corvus  frugilegus,  Linnaeus.    The  Rook. 

On  the  Moorish  side  of  the  Straits  I  can  find  no  record  of  the 
occurrence  of  the  Rook.  I  never  met  with  any  near  Gibraltar, 
or,  indeed,  further  south  than  the  Goto  del  Rey,  in  the  neigh- 
bourhood of  Seville,  where  there  were  several  large  flocks  in 
January  ;  and  they  appear  to  be  there  regular  winter  visitants. 

Black,  with  blue  gloss. 

Adults.  With  bare  scurfy  skin  on  forehead,  lores,  and  throat. 

Young  in  first  plumage  have  these  parts  feathered,  and  resemble  young 
Carrion-Crows,  but  the  bases  to  the  body-feathers  are  grey  and  the  bill  is 
slender.  Length  18-20  inches. 


LANIID.E.  87 


L  A  N 1 1 D  $1 .     Young  barred  below. 

84.  Lanius   meridionalis,   Temminck.     The    Spanish    Grey 

Shrike. 

Spanish.  Alcaudon  real. 

This  Shrike  is  scarce  in  the  neighbourhood  of  Gibraltar,  the 
few  specimens  I  have  seen  there  having  occurred  in  autumn  only. 
Further  north,  though  rather  local,  they  become  common  in  many 
places,  being  abundant  and  resident  in  the  scrubby  jungle  near 
Seville.  They  nest,  about  the  15th  of  April,  in  bushes  and  low 
trees,  building  a  large  nest,  the  internal  diameter  measuring  some 
five  inches.  When  placed  on  a  bough,  the  lower  half  is  some- 
times made  of  mud,  the  upper  half  being  constructed  with  rough 
grass  lined  with  fine  grass,  the  whole  covered  outside  with 
lichens  and  bits  of  Cudweed  (GnaphaUum  luteo-album] — the  same 
plant  so  much  used  by  the  Woodchat  and  many  other  birds. 
The  eggs  are  from  four  to  five  in  number.  At  one  time  I  was 
under  the  erroneous  impression  that  these  birds  were  migratory, 
from  having  seen  one  or  two  near  Gibraltar  in  autumn;  but, 
never  having  seen  or  heard  of  them  on  the  African  side,  must 
have  been  mistaken ;  besides,  they  were  more  numerous  about 
the  Goto  del  Key  in  winter  than  in  May.  However,  this  tends 
to  show  that  they  shift  their  ground  in  Spain,  though  not 
migrating  out  of  the  country. 

Sexes  alike.  Above  bluish  grey  ;  below  with  rosy  tinge,  except  the  grey 
flanks ;  narrow  white  stripe  from  forehead  over  eye.  Length  9  inches. 

85.  Lanius  algeriensis,  Lesson.    The  Algerian  Grey  Shrike. 

Moorish.  Bou-serund  (Favier). 

Favier's  remarks,  which  under  the   synonym  "meridionalis" 
refer  to  this  species,  do  not  add  anything  to  the  following  notes. 
This  Shrike,  which  would  be  more  fitly  named  the  North-west- 


88  LANIIDJ3. 

African  Shrike,  as  it  is  not  peculiar  to  Algeria,  is  not  common  in 
the  immediate  vicinity  of  Tangier ;  but  a  little  further  south, 
near  Larache,  and  towards  the  Fondak  on  the  road  to  Tetuan 
they  are  very  abundant.  Their  habits,  nests,  and  eggs  are 
identical  with  those  of  L.  meridionalis,  their  Spanish  repre- 
sentative, and  they  are  only  met  with  in  scrubby  jungle.  On  the 
18th  of  April  we  found  one  nest  in  a  thick  lentiscus  bush,  with 
five  eggs,  which  were  hard  sat-on,  and  another  on  the  20th,  with 
only  three  fresh  eggs. 

This  species  is  distinguishable  from  the  Spanish  Grey  Shrike  by  the 
grey  colour  of  the  underparts. 

Plumage  black,  white,  and  grey. 

Above  dark  bluish  grey  ;  below  grey  ;  no  white  streak  over  eye.  Length 
9^  inches. 

86.  Lanius  minor,  J.  F.  Grmelin.    The  Lesser  Grey  Shrike. 

This  small  migratory  Shrike  is  said  to  have  been  obtained  at 
Lanjaron  (Arevalo],  but  Lord  Lilford  informs  me  (1894)  that  he 
never  met  with  it  in  Andalucia  and  that  his  record  of  the  bird 
there  was  an  error.  It  is  reported  to  occur  in  the  east  of  Spain. 

First  primary  very  short,  less  than  one  third  of  second  ;  tail  even. 
Male.  Above  grey  ;  forehead  and  feathers  from  bill,  round  eye  to  ear- 
coverts,  black  ;  breast  and  flanks  with  a  roseate  blush.     Wing-bar  white. 
female.  Less  black  on  fore  part  of  head. 
Young.  Has  only  black  round  and  behind  eye.     Length  8-9  inches. 

87.  Lanius  collurio,  Linnseus.      The   Red-backed   Shrike  or 

"  Butcher-bird." 

Mr.  Meade- Waldo  observed  this  Shrike  near  Tangier  on  the 
10th  of  April,  1892,  and  it  has  been  obtained  there,  though 
very  rarely  in  Andalucia ;  so  we  may  reasonably  assume  that 
their  line  of  migration  is  not  across  the  Straits  of  Gibraltar ;  and 
it  would  be  interesting  to  know  the  reason  of  this,  as  in  187G 
Lord  Lilford  and  myself  found  them  common  in  the  Province  of 
Santander. 


LANIUS  POMERANUS.  89 

Male.  Scapulars  dull  brick-red,  uniform  with  the  back  ;  crown,  hind 
neck,  and  rump  grey  ;  chin  white  ;  breast  with  roseate  tinge. 

Female.  Above  brownish  grey  ;  tail  brownish  red.  Underparts  white  ; 
sides  of  neck,  breast,  and  flanks  barred  with  brown. 

Young.  Above  reddish  brown  ;  below  pale  brown  ;  above  and  below 
barred  with  blackish  brown.  Length  7^-8  inches. 

88.  Lanius  pomeranus,  Sparrman.    The  "Woodchat  Shrike. 

Moorish.  Aisha  el  kra  (Favier)  *.     Spanish.  Alcaudon. 

As  common  in  Morocco  as  in  Andalucia,  the  Woodchat  arrives 
in  March  and  April,  leaving  in  August  and  September.  The 
first  arrival  noticed  at  Gibraltar  in  1868  was  on  the  3rd  of  April, 
in  1869  on  the  3rd  of  April,  in  1870  on  the  29th  of  March,  in 
1894  on  the  28th  of  March ;  the  passage  ceases  about  the  20th 
of  April.  The  latest  seen  was  on  the  14th  of  October,  1871 ;  in 
1869  I  observed  them  returning  south  on  the  26th  of  August. 

The  Woodcliat  is  one  of  the  most  abundant  and  conspicuous 
birds  in  spring  on  both  sides  of  the  Straits.  Very  tame  and 
confiding,  unlike  their  big  cousins  L.  meridionalis  and  L.  alye- 
riensis ;  their  pied  appearance  and  the  bright  chestnut-coloured 
head  of  the  adult  males  cause  them  to  be  noticed  even  by  the 
most  unobservant.  They  are  to  be  seen  in  every  direction  in 
woods  and  on  plains,  perched  on  tops  of  trees,  bushes,  aloes,  and 
tall  plants,  making  their  larders  on  the  spikes  of  the  aloes,  and 
impaling  on  the  thorns,  beetles,  bees,  and  all  kinds  of  insects, 
and  are  extremely  mischievous  among  bees. 

The  nest,  which  usually  contains  eggs  by  the  10th  of  May, 
is  a  small  edition  of  that  of  L.  meridionalis,  but  more  covered 
outside  with  the  greyish-white  flowers  and  stalks  of  Cudweed 
(Gnaphalium  luteo-album),  and  usually  placed  low  down  in  trees. 

*  "  Kra  "  is  a  kind  of  ringworm,  and  no  doubt  given  as  a  name  to  the  Woodchat 
owing  to  the  supposed  resemblance  the  chestnut  head  of  the  males  has  to  the 
head  of  a  Moor  afflicted  with  this  disorder,  which  is  common  in  Tangier,  and 
causes  a  rusty  mangy  baldness.  "Aisha  "  is  a  female  name  :  why  applied  to  pied 
birds  ? 


90  LANIID^E. 

The  eggs,  four  or  five  in  number,  sometimes  six,  are  subject  to 
great  variation,  many  resembling  those  of  our  English  Butcher- 
bird, L.  collurio. 

Male.  Forehead  black  ;  centre  of  crown  and  nape  chestnut ;  scapulars 
white,  forming  a  large  white  patch  on  each  side  ;  rump  and  underparts 
white  ;  white  alar  bar,  or  speculum,  formed  by  the  white  outer  web  of  base 
of  primaries. 

Female.  Like  the  male,  but  markings  ill-defined  and  not  so  bright. 

Young.  Much  like  young  of  last,  but  lighter.  Scapulars  and  rump  paler  ; 
plumage  vermiculated  with  dark  brown.  Length  7^  inches. 

89.  Telephonus  erythropterus  (Shaw).     The  Hooded  Shrike 
or  Tchagra. 

Moorish.  Abermat  (Favier). 

"  This  Shrike  is  nearly  as  common  as  Lanius  algeriensis  near 
Tangier,  and  is  resident,  nesting  in  bushes  twice  a  year — in  May 
or  June,  and  again  in  November.  Their  note,  which  is  a  kind 
of  whistle,  harmonious  and  well  sustained,  and  very  like  that  of 
a  Blackbird,  is  usually  beard  from  the  middle  of  some  thick  bush 
(where  they  have  a  habit  of  hiding  themselves),  as  well  as  in  the 
thickest  part  of  trees.  They  lay  about  three  eggs,  of  the  same 
shape  as  those  of  other  Shrikes,  but  white  marked  with  lines  and 
small  spots  of  ash-brown  and  russet,  mostly  at  the  thick  end. 
The  sexes  are  alike  in  plumage,  and  undergo  no  change."- 
Favier. 

According  to  my  limited  experience,  the  Tchagra  is  rather 
scarce  near  Tangier,  but  more  plentiful  about  a  day's  journey 
south.  On  the  25th  of  April  we  took  a  nest  in  a  small  tree  close 
to  the  ground,  containing  three  slightly  incubated  eggs,  which 
is,  we  were  informed,  the  usual  complement.  ,  The  nest  was  not 
so  compact  as  that  of  the  Woodchat,  containing  less  grass  and 
dried  flowers,  being  chiefly  built  of  fibrous  roots.  Mr.  Meade- 
Waldo  writes  that  "  when  beating  for  pigs,  and  these  birds  are 
hustled,  they  dart  into  a  thick  bush  and  sit  perfectly  still,  so  that 
you  can  look  at  them  from  a  yard's  distance  or  less ;  when 


MUSCICAPIDJS.  91 

undisturbed  in  the  early  morning,  they  have  a  most  beautiful 
wild  rich  song."  It  is  an  easy  bird  to  recognize  when  once  seen 
on  the  wing,  the  chestnut  of  the  wing-coverts  and  the  long  tail 
being  very  conspicuous. 

Favier  states  that  they  cross  the  Straits  ;  all  I  can  say  is,  I  never 
saw  a  Spanish  specimen :  if  they  do  occur  in  Spain,  they  are  as 
yet  unobserved  by  any  competent  ornithologist. 

Crown  black  ;  broad  white  eye-stripe,  with  black  stripe  below. 

Wing-coverts  chestnut ;  below  dark  ash-grey.  Tail  much  graduated, 
all  the  feathers,  except  the  two  light  brown  centre  ones,  black,  tipped  with 
white.  Length  9  inches,  tail  4  inches. 


MUSCICAPID^B.     Nostrils  more  or  less  covered  by 

bristly  hairs. 

90.  Muscicapa  grisola  (Linnaeus).    The  Spotted  Flycatcher. 

Spanish.  Papamoscas,  Pifiata. 

"  This  Flycatcher  is  very  common  near  Tangier,  where  they 
arrive  in  April  and  May  in  pairs  and  small  nights,  some  remaining 
to  nest,  the  others  passing  across  the  Straits  to  return  in 
September,  when  they  disappear.  Near  Rabat  they  are  called 
Sorsh  by  the  Arabs." — Favier. 

The  Spotted  Flycatcher  is  exceedingly  numerous  near  Gibraltar, 
chiefly  nesting  in  the  pine- woods,  and  was  first  seen  on  the  llth 
of  May,  1870,  on  the  3rd  of  the  same  month  in  1871,  and  on  the 
8th  in  1874.  The  first  egg  obtained  was  on  the  24th  of  the  same 
month.  Verner  found  a  nest  with  fully-fledged  young  on  3rd 
of  July,  1879.  I  regret  not  to  have  any  note  of  their  departure  ; 
but  it  is  previous  to  the  middle  of  September,  and  Verner  informs 
me  that  they  leave  about  the  end  of  August. 

Ashy  brown  above  ;  forehead  lighter  than  head,  which  is  mottled  with 
streaks  of  darker  brown  ;  breast  white,  with  brown  streaks  ;  axillaries 
fawn-coloured. 

Young.  Spotted.     Length  5£  inches. 


92  MUSCICAPID^E. 

91.  Muscicapa  atricapilla,  Linnaeus.    The  Pied  Flycatcher. 

Spanish.  Cerrojillo. 

"Very  abundant  near  Tangier  on  passage,  crossing  to  Europe 
in  pairs  and  small  nights  during  April  and  May,  returning  in 
September  and  October." — Favier. 

In  Andalucia  the  Pied  Flycatcher  only  appears  during 
migration ;  and  I  was  never  able  to  detect  them  remaining  to 
nest.  The  earliest  date  of  arrival  noticed  was  the  8th  of  April  ; 
from  then  till  the  1st  of  May  they  pass  in  great  numbers, 
returning  late  in  September.  The  latest  date  on  which  they 
were  observed  was  the  17th  of  October,  1870,  when  I  found  one 
in  an  owl's  disused  cage,  where  there  was  a  lot  of  carrion  which 
attracted  flies  ;  and,  again,  in  1871  I  noticed  them  on  the  16th 
of  October. 

Male  in  spring.  Above  black,  except  white  forehead  and  white  wing-bar  ; 
axillaries  and  underparts  white  ;  quills  brownish  black.  Tail  black,  three 
outer  feathers  on  each  side  having  two-thirds  of  the  outer  web  from  the 
base  white. 

Female,  male  in  winter,  and  young.  Above  brownish  grey  ;  underparts 
dirty  white.  Tail  as  in  male,  but  duller.  Length  5  inches. 


92.  Muscicapa     collaris,    Bechstein.        The    Collared    Fly  - 
catcher. 

This  Flycatcher  is  recorded  by  Mr.  Saunders  as  having  been 
once  seen  by  him  at  Seville.  Is  said  to  be  common  in  Portugal, 
therefore  it  is  somewhat  singular  that  the  bird  should  not  be 
more  frequently  noticed  in  Andalucia. 

The  adult  male  differs  only  from  M.  atricapilla  in  having  a  white  collar 
on  the  back  of  the  neck. 


HIRUNDIN1D.E.  93 

HIRUNDINID^. 

93.  Hirundo  rustica,  Liunseus.    The  Common  Swallow. 

Moorish.  Khotaifa.     Spanish.  Golondrina. 

"  Great  flights  of  Swallows  pass  in  January  and  February  to 
Europe,  returning  in  September  and  October  to  join  those  which 
remain  near  Tangier  to  nest,  all  leaving  to  go  further  south  for 
the  winter.  The  Moors  believe  that  it  oifends  God  to  kill  these 
birds,  in  the  same  way  as  they  believe  it  pleases  or  soothes  the 
Evil  One  to  kill  the  Raven,  The  stories  on  which  this  supersti- 
tion is  founded  are  too  long  to  relate ;  but  I  was  informed  by 
one  person  that  the  Swallows  and  White  Storks  were  inspired  by 
Allah  to  protect  the  harvest  and  the  country  from  noxious  insects 
and  reptiles,  and  that  the  birds  themselves  (knowing  the  benefits 
they  confer  on  man)  ask  in  return  protection  for  their  offspring 
by  building  their  nests  on  the  walls  of  towns  and  houses,  and 
that  therefore  any  one  who  kills  them  must  be  a  Kaffir,  i.  e.  not 
a  true  believer  of  the  Prophet,  especially  as  the  birds  would  only 
be  killed  for  mischief,  being  useless  when  dead." — Favier. 

I  wish  this  belief  could  be  instilled  into  the  minds  of  English 
people,  who  kill  and  destroy  every  rare  bird  they  see,  through 
ignorance,  love  of  destruction,  and  to  gratify  the  cupidity  of 
private  collectors. 

About  Gibraltar  the  Swallow  generally  arrives  about  the  13th 
of  February,  although  a  straggler  is  occasionally  seen  in  December 
and  January.  I  have  observed  them  crossing  the  Straits  in 
considerable  numbers  up  to  the  15th  of  April;  the  latest  passing 
were  noticed  on  the  24th  of  that  month.  Their  migration,  like 
the  nest  of  the  Hirundinida?,  is  spread  over  a  considerable  period ; 
I  have  observed  the  nest  finished  on  the  23rd  of  February,  and 
young  birds  able  to  fly  on  the  24th  of  May.  One  of  each  pair, 
when  they  first  arrive,  is  tinged  with  a  rufous-buff  colour  on  the 
underparts ;  and  as  these  are  slightly  larger  in  size,  I  think  they 
are  the  male  birds ;  but  did  not,  even  for  the  sake  of  proving 


94  HIRUNDIN1DJS. 

this,  kill  one.  I  remember,  on  a  very  cold  day  (the  13th  of 
March,  1874),  Mr.  Stark  particularly  drawing  my  attention  to 
this  difference  in  the  pairs  of  birds,  which,  driven  by  the  cold 
into  the  stables  and  outhouses  of  the  venta  at  Pulverilla,  were 
sitting  side  by  side,  touching  one  another,  allowing  us  almost  to 
touch  them.  The  contrast  in  their  colour  was  then  most  conspi- 
cuous ;  but  they  appear  gradually  to  lose  this  rufous  tinge  as  the 
season  advances,  and  by  the  end  of  April  it  is  not  apparent. 

In  the  month  of  March,  during  the  time  of  the  northern 
migration  of  Swallows,  a  small  butterfly,  Thestor  ballus  (one  of 
the  Lycsenidse,  between  the  Hair-Streaks  and  the  Coppers),  is  out 
in  great  abundance  on  the  plains.  When  walking  across  the 
grass,  the  Swallows,  which  keep  flying  very  close  to  leeward  of  you, 
instantly  catch  any  unfortunate  ballus  that  flies  up  ;  but  they  seem 
unable  to  take  them  on  the  ground,  perhaps  from  the  protective 
colouring  of  their  green  under-wings  they  cannot  see  them  when 
at  rest ;  but  any  way  the  Swallow  is  an  annoyance  to  the  butterfly 
collector. 

Head,  back,  and  rump  steel-blue.  Throat  and  forehead  chestnut-red, 
with  complete  blue-black  band  across  the  lower  part  of  the  throat ;  under- 
parts  buff  to  creamy  white.  Tail  strongly  forked  ;  two  outer  feathers 
greatly  prolonged.  Length  8^  inches. 

94.  Hirundo  rufula,  Temminck.    Red-rumped  Swallow. 

Arevalo  mentions  this  Eastern  Swallow  as  having  nested  near 
Malaga,  and  it  is  also  recorded  from  Valencia. 

Above  glossy  purple-blue,  with  whitish  streaks  on  back.  Below  white, 
tinged  with  buff  and  streaked  with  dark  brown. 

O 

Nape  and  stripe  over  eye  (forming  a  broad  collar)  and  rump  brick-red. 
Lower  part  of  tail-coverts  white.  Length  7^  inches. 

95.  Chelidon  urbica  (Linnaeus).    The  House-Martin. 

Spanish.  Vencejo. 

"  As  common  as  the  Swallow  near  Tangier,  this  species  is  seen 
in  flights  on  passage,  crossing  to  Europe  in  February,  returning 


COTILE  EIPAR1A.  95 

in  September  and  October,  frequently  travelling  in  company  with 
Hirundo  rustica,  and,  as  in  their  case,  remaining  to  breed  in 
some  numbers.  They  often  make  their  nests  touching  one 
another,  as  many  as  sixty  being  joined  together ;  the  entrance- 
hole  is  sometimes  at  the  side,  sometimes  in  the  centre,  according 
to  the  position  of  the  nest.  They  are  named  '  Khotaifa '  by  the 
Arabs,  indiscriminately  with  the  Swallow." — Favier. 

The  above  notes  equally  apply  to  the  House-Martin  in  Anda- 
lucia.  The  earliest  date  of  arrival  noticed  at  Gibraltar  was  the 
5th  of  February.  Verner  gives  notes  of  great  numbers  passing 
on  15th  of  April,  1875,  and  many  passing  on  the  17th  of  February, 
1879. 

Both  this  bird  and  the  Swallow  frequently  nest  in  caves  and 
on  overhung  rocks. 

Bump  white ;  under  surface  pure  white  ;  tail  much  forked ;  toes 
feathered.  Length  4f  inches. 

96.  Cotile  riparia  (Linnseus).    The  Sand-Martin. 

"  Migratory,  and  the  least  abundant  of  the  Swallows  about 
Tangier,  arriving  to  cross  the  Straits  in  March  and  April,  returning 
in  October  to  disappear  for  the  winter." — Favier. 

We  found  the  Sand-Martin  at  Has  el  Doura,  in  Morocco,  in 
small  numbers,  and  have  no  doubt  that  they  were  nesting  in  the 
vicinity.  They  breed  in  the  neighbourhood  of  Seville,  but  near 
Gibraltar  are  only  met  with  on  passage.  The  first  seen  by  me 
was  on  the  24th  of  March  1870,  22nd  of  March  1871,  24th  of 
February  1872,  28th  of  February  1874;  they  were  seen  passing 
as  late  as  the  24th  of  April.  On  the  loth  of  May  I  saw,  in  the 
evening,  over  some  marshes  near  Vejer,  a  flight  of  Sand-Martins 
numbering  many  hundreds — I  might  say,  thousands.  I  noticed 
them  on  the  14th  of  October  on  their  southward  journey. 

Above  mouse-coloured  ;  underparts  white,  except  greyish  band  between 
chest  and  throat.  Minute  tuft  of  feathers  just  above  hind  toe. 

Young.  Feathers  above  edged  with  huffish  white.     Length  4f  inches. 


96  HIRUNDINID.E. 

97.  Cotile  rupestris  (Scopoli).    The  Crag-Martin. 

Spanish.  Vencejillo. 

"  Nearly  as  common  as  the  House-Martin  about  Tangier. 
Sometimes  they  pass  in  large  flights,  crossing  the  Straits  in 
February  and  March,  returning  in  October  and  November." — 
Favier. 

The  Crag-Martin,  though  universally  distributed  during  the 
breeding-season  in  the  rocky  sierras,  is  to  a  great  extent  migra- 
tory. Those  which  do  not  quit  the  country  appear  during  the 
daytime  in  low  ground  near  the  coast  about  the  middle  of 
October,  great  numbers  being  then  seen  at  Gibraltar.  In  March 
they  return  to  their  breeding-haunts,  some  nesting  in  inaccessible 
places  at  the  "  back  of  the  Rock." 

They  commence  about  the  10th  of  March  to  build  their  nests, 
which  resemble  those  of  the  Swallow,  H.  rustica,  but  being  placed 
on  the  roofs  of  caverns  are  very  difficult  to  reach ;  and  I  did  not 
succeed  in  examining  the  inside  of  one.  The  birds  were  sitting 
by  the  30th  of  April.  One  locality  for  nests  near  Gibraltar,  and 
the  most  accessible  that  I  have  seen,  was  a  cave  in  a  patch  of 
rocks  at  the  entrance  of  La  Trocha,  on  the  road  from  Algeciraz 
to  Ojen,  where  it  passes  by  the  side  of  the  ravine  called  la 
Garganta  del  Capitan. 

At  the  back  of  the  Rock,  at  Gibraltar,  is  a  cave  almost  under 
the  Osprey's  eyry,  which  can  only  be  entered  by  landing  from  a 
boat  in  fair  weather.  This  cave  is  very  large  and  open,  with 
sand  at  the  bottom  sloping  upwards  for  a  considerable  distance  at 
a  sharp  angle,  and  at  the  end,  judging  from  the  tracks  of  divers 
Genets  or  Striped  Cats  (Viuerra),  seems  to  be  the  regular  dining- 
room  of  these  animals  ;  for  whenever  we  visited  the  place  it  was 
covered  with  the  tail-feathers  and  pinions  of  numbers  of  Rock- 
Martins  mingled  with  those  of  a  good  many  Swifts,  Rock-Doves, 
and  a  few  Lesser  Kestrels. 

Above  mouse-coloured  ;  underparts  huffish  brown.  Oval  white  spots  on 
all  but  the  centre  and  outer  tail-feathers.  Length  5  inches. 


FRINGILLID^E.  97 

FRINGILLIDJE. 

98.  Coccothraustes  vulgaris,  Pallas.    The  Hawfinch. 

Spanish.  Cascanueces  (Nutcracker),  Pinonero,  Pico  gordo. 

Favier  states  the  Hawfinch  to  be  "  very  rare  near  Tangier, 
having  only  met  with  two — one  in  1836,  the  other  in  1849." 
Olcese  obtained  one  since  then  (Reid,  '  Ibis,'  1885),  whilst 
Mr.  Meade- Waldo  saw  several  near  Tangier  in  the  spring 
of  1892. 

On  the  Spanish  side  of  the  Straits  this  bird  is  very  common, 
and  most  plentiful  in  winter.  Some  nest  in  the  Cork-wood 
in  May ;  and  during  the  season  of  migration  they  often 
frequent  pine-woods,  and  are  then  rather  shy  and  difficult  to 
approach.  About  Cordova  they  are  most  abundant,  and  are 
there  and  at  Seville  exposed  alive  for  sale  at  about  one  real 
apiece.  I  kept  a  pair,  purchased  at  Seville,  for  some  time ; 
but  never  could  tame  them.  The  hen  bird  at  last  killed  her 
mate,  having  previously  at  regular  intervals  plucked  him  while 
living. 

I  gave  this  amiable  and  domestic  female  to  a  bird-fancier  at 
Gibraltar,  much  to  his  delight,  but  ultimately  to  his  sorrow,  as 
she  vented  her  temper  upon  some  other  pet  birds  with  which 
she  was  caged,  and,  in  consequence,  justly  suffered  capital 
punishment. 

Bill  very  large.     Fifth  and  next  four  primaries  shaped  like  a  bill-hook. 

Adult  male.  General  colour  above  chestnut-brown ;  crown  reddish 
brown,  lightest  on  forehead ;  hind  neck  grey  ;  black  patch  on  chin  ;  larger 
wing-coverts  tipped  with  white  ;  sides  of  body  and  flanks  vinaceous  brown  ; 
bill  bluish  ;  iris  greyish  white. 

Female.  Head  ashy  brown  ;  secondaries  edged  with  bluish  grey  ;  flanks 
as  in  male. 

In  winter.  Bill  fleshy  white. 

Young.  Spotted,  no  grey  at  back  of  neck  or  black  patch  on  chin  ;  lower 
breast  and  flanks  spotted  and  barred  with  brown  ;  iris  brown.  Length 
7  inches. 

II 


98  FRINGILLID^E. 

99.  Ligurinus  chloris  (Linnaeus).    The  Greenfinch. 

Spanish.  Verdon. 

"Found  near  Tangier  as  a  common  resident;  others  migrate 
in  immense  flights,  which  pass  north  in  February  and  March, 
returning  in  October  and  November." — Famer. 

This  species,  another  of  our  common  British  birds,  is  extremely 
abundant  on  both  sides  of  the  Straits.  Many  are  resident, 
nesting  during  the  month  of  May ;  and  hundreds  are  caught 
in  August  and  September  and  brought  into  the  markets,  where 
they  are  exposed  for  sale  in  large  bunches.  The  Greenfinch 
is  also  a  very  common  cage-bird,  for  sometimes  as  many  as 
twenty  may  be  seen,  each  in  a  separate  cage,  hanging  outside 
the  wall  of  a  house.  What  their  merits  as  a  cage-bird  are  it  is 
difficult  to  understand,  as  their  song,  if  it  can  be  dignified  with 
the  name,  is  to  me  positively  unpleasant. 

Greenfinches  from  Morocco  and  the  south  of  Spain  are  rather 
smaller  and  more  brightly  coloured  than  English  birds. 

Adult  male.  General  colour  above  olive -yellow,  shaded  with  ashy  grey  ; 
crown  more  ashy  than  back  ;  axillaries  bright  yellow  ;  primaries  edged 
•with  bright  yellow  reaching  to  the  shaft  ;  tail  black  at  ends,  yellow  at  the 
base  ;  outer  tail-feathers  yellow  at  the  base  of  both  webs. 

In  winter.  Browner. 

Female.  Browner  than  male;  primaries  only  margined  with  yellow; 
outer  tail-feathers  edged  with  yellow  on  outer  web  only. 

Young  male.  Like  female,  but  has  primaries  and  tail  as  in  adult  male. 
Length  (j  inches. 

100.  Carduelis  elegans,  Stephens.    The  Goldfinch. 

Moorish.  Mouknin.  Spanish.  Gilguero.  Andalucian.  Sil- 
guero. 

"Exceedingly  plentiful  near  Tangier,  and  resident;  but 
numbers  migrate,  arriving  from  about  the  month  of  August, 
to  depart  again  for  the  north  in  the  month  of  March." — Famer. 

The   Goldfinch   is,    without    doubt,    the    most    common    and 


CARDUELIS  ELEOANS.  99 

abundant  bird  in  the  west  of  Andalucia.  Always  plentiful  in 
every  direction,  they  appear  in  countless  flocks  when  the  seed 
of  the  various  thistles  becomes  ripe  ;  and  Spain  is  the  country 
par  excellence  both  of  thistles  and  donkeys.  The  former,  in 
some  of  the  vegas  and  plains,  grow  in  almost  impenetrable 
thickets,  in  places  covering  acres  of  ground ;  for  when  the  land 
is  left  fallow  for  a  season,  all  weeds  are  allowed  to  run  riot, 
and  they  do  so  with  a  vengeance.  Some  of  the  thistles  (and 
there  are  many  different  kinds)  are  very  handsome — a  large, 
yellow,  carline  species  being  perhaps  the  most  attractive  to  the 
eye.  The  stalks,  heads,  and  leaves  of  a  species  of  Cynara,  very 
like  the  garden  artichoke,  are  a  staple  vegetable  (cardo]  with  the 
Spaniards.  The  dried  stalks  of  another  thistle  are  much  used 
as  tinder  in  the  rural  districts,  and  known  as  yesca  de  cardo, 
which  takes  light  well  from  the  sparks  made  by  a  flint  and 
steel,  most  of  the  peasants  using  no  other  method  of  lighting 
their  papelitos. 

To  return  to  the  Goldfinches,  at  the  time  of  their  thistle- 
harvest  they  are  caught  in  vast  quantities  in  clap-nets ;  and  it 
is  not  unusual  to  see  a  man  with  bunches  of  several  hundreds, 
which  are  sold  at  a  ridiculously  low  price.  Though  perhaps 
rather  dry,  they  are  not  to  be  despised  as  morsels ;  but  one 
feels  as  if  committing  a  sin  when  devouring  such  a  charming 
and  useful  little  bird. 

The  Goldfinch  in  Andalucia  breeds  about  the  beginning  of 
May,  and  occasionally  nests  at  Gibraltar  in  the  Alameda  and 
various  gardens  at  the  South. 

Bill  nearly  conical  and  sharp-pointed. 

Adult.  General  colour  above  ruddy  brown.  Forehead,  upper  throat, 
and  cheeks  crimson  ;  hinder  crown  and  sides  of  neck  black  ;  outer  web 
of  basal  half  of  primaries,  except  the  first,  and  the  larger  wing-coverts, 
brilliant  yellow. 

Young.  Mottled,  no  red  or  black  on  head,  that  part  being  greyish  brown. 
Length  5  inches. 

H2 


100  FEING1LLID.E. 

101.  Carduelis  spinus  (Linnaeus).    The  Siskin. 

Spanish.  Lugano. 

On  the  African  side  of  the  Straits,  Mr.  Meade-Waldo  found 
the  Siskin,  in  the  early  spring  of  1892,  abundant  near  Tangier. 
Neither  Favier  nor  Mr.  Drake  mention  the  bird. 

In  Andalucia  they  are  very  irregular  in  appearance,  in  some 
winters  not  being  noticed  at  all.  The  Spaniards  say  they  only 
come  every  seventh  year.  This  idea  is  prevalent  about  Seville, 
as  well  as  near  Gibraltar ;  but,  it  is  needless  to  say,  is  a  popular 
error. 

In  the  winter  of  1870-71  they  were  plentiful  wherever  there 
were  any  alder  trees ;  and  I  saw  some  as  late  as  the  4th  of 
April.  In  the  two  previous  winters,  and  during  the  one 
following,  none  were  obtained  by  the  birdcatchers,  who  are 
always  looking  out  for  them,  as  they  are  much  desired  and 
fetch  a  good  price  as  cage-birds.  In  1874,  I  saw  four  on  a 
white  poplar  tree  in  the  Alameda  of  Gibraltar,  on  the  24th  of 
March  ;  they  were  so  tame  as  to  allow  of  my  approach  within 
a  yard  of  them,  and  remained  for  a  long  time  close  to  me. 

Adult  male.  Above  yellowish  green  ;  cheeks,  throat,  and  breast  bright 
yellow  ;  flanks  streaked  with  black  ;  crown  and  small  patch  on  chin 
black  ;  tail  blackish,  all  but  the  two  centre  feathers  yellow  at  the  base. 

Female.  No  black  on  head  or  chin  ;  throat  uniform  dull  white. 
Length  4|  inches. 

102.  Chrysomitris  citrinella  (Linnseus).    The  Citril  Finch. 

Reported  by  Mr.  Saunders  to  be  found  on  the  coast,  and 
Arevalo  says  they  are  of  irregular  occurrence  in  autumn  near 
Granada  and  Malaga.  I  never  met  with  any  near  Gibraltar, 
and  Lord  Lilford  informs  me  that  he  never  handled  a  Spanish 
specimen;  while  Major  Verner,  who  knew  the  bird  well,  tried 
in  vain  to  obtain  or  even  see  one  during  six  years  at  Gibraltar, 
none  even  being  brought  into  the  market. 

Male.  Back  dull  olive-green,  each  feather   striped  with   dark   brown. 


PASSER  DOMESTICUS.  101 

Crown,  cheeks,  rump,  and   underparts   greenish  yellow.     Nape,  hind  neck, 
sides  of  neck,  and  flanks  ashy  grey. 

Female.  Duller  in  plumage.     Length  5  inches. 


103.  Serinus  hortulans,  K.  L.  Koch.    The  Serin  Finch. 
Spanish.  Chamaris. 

"  This  bird  is  very  abundant  near  Tangier,  both  as  a  resident 
and  on  migration,  when  they  are  seen  passing  north  in  immense 
flights  during  February,  returning  in  October  and  November." 
— Favier. 

The  Serin  Finch  is  found  on  the  Spanish  side  in  accordance 
with  the  above  note.  During  the  breeding-season  they  greatly 
frequent  the  Cork-wood,  and  their  hissing  unpleasant  song  is 
to  be  heard  all  around.  They  seem  to  keep  very  much  to  the 
banks  of  rivers,  nesting  in  May  on  trees  and  bushes,  like  the 
Goldfinch,  resembling  that  bird  both  in  their  nest  and  eggs. 

Bill  conical,  short,  stout,  and  blunt. 

Adult  male.  General  colour  above  pale  brown,  each  feather  edged 
with  yellow,  and  streaked  with  black  on  the  centre ;  crown  like  the 
back ;  forehead,  nape,  eye-stripe,  throat,  and  breast  yellow ;  flanks  streaked 
with  black. 

Female.  Like  male,  but  duller  ;  throat,  breast,  and  flanks  streaked  with 
dark  brown.  Length  4^  inches. 

104.  Passer  domesticus  (Linnseus).    The  House-Sparrow. 

Moorish.  Bertal.     Spanish.  Gorrion. 

Common  on  both  sides  of  the  Straits,  being  the  Sparrow  of 
the  district. 

Adult  male.  Crown,  nape,  and  rump  ashy  grey,  with  a  chestnut  streak 
on  each  side  of  head ;  throat  black  ;  ear-coverts  white  j  general  colour 
above  chestnut,  streaked  with  black  on  back. 

Female.  Lacks  black  throat  of  male  ;  above  dingy  brown,  streaked  with 
black ;  pale  eye-stripe  ;  rump  ashy  brown. 

Young.  Like  female,  but  can  always  be  known  by  the  yellowish  skin  at 
angle  of  gape.  Length  6  inches. 


102  FRINGILL1D.E. 

105.  Passer  montanus  (Linnaeus).    The  Tree-Sparrow. 
Spanish.  Gorrion  serrano. 

This  Sparrow  occurs  sometimes  in  Andalucia,  as  I  have  seen 
specimens  obtained  in  the  country ;  but  I  did  not  observe  any 
near  Gibraltar. 

Sexes  alike.  Smaller  than  last.  Always  nests  in  holes  of  trees,  thatch, 
and  buildings. 

Adult.  Crown  and  nape  uniform  vinaceous  chestnut ;  throat  black ; 
ear-coverts  ashy  white,  with  black  patch  on  lower  part  ;  rump  ashy  brown. 

Young.  Duller,  but  show  indistinctly  markings  of  adults.  Length 
5^  inches. 

106.  Passer  hispaniolensis,  Temminck.   The  Spanish  Sparrow. 

This  is  another  of  the  chestnut-headed  Sparrows,  and  is  local 
in  distribution  on  both  sides  of  the  Straits.  In  some  places 
they  are  very  abundant ;  and,  as  is  well  known,  often  build  under 
the  nests  of  the  larger  birds  of  prey.  I  found  a  nest  built 
underneath  one  of  Buteo  desertorum. 

Male.  Crown  and  nape  chestnut  ;  cheeks  whitish  ;  back  black,  each 
feather  streaked  with  white  ;  flanks  streaked  with  black  ;  throat  and  chest 
black. 

Females  resemble  those  of  the  Common  Sparrow.     Length  5^  inches. 

107.  Petronia  stulta  (Gmelin).    The  Rock-Sparrow. 

Spanish.  Gorrion  montes. 

Neither  Favier  nor  Mr.  Drake  mentions  having  seen  this 
Sparrow  in  Morocco,  where,  however,  it  is  found,  as  on  the 
Spanish  side,  commonly  in  the  sierras  and  rocky  ground,  nesting 
in  May  in  holes  of  rocks. 

The  adult  male  has  a  yellow  spot  on  the  throat ;  in  the  female  this  mark 
is  much  fainter,  and  absent  in  the  young.  Length  6  inches. 

108.  Fringilla  ccelebs,  Linnams.    The  European  Chaffinch. 

Spanish.  Pinzon. 

This  Chaffinch  crosses  in  winter  to  the  African  side  of  the 


FRINGILLA.  SPODIOGENTS.  103 

Straits,  as  Capt.  S.  Reid  in  1884-85  found  many  in  the  Tzelatza 
Valley  ;  whilst  Mr.  Meade- Waldo  informs  me  that  in  the  spring 
of  1892  he  saw  immense  numbers,  far  more  than  of  the  next 
species.  Throughout  Andalucia  where  there  are  trees  this 
bird  is  as  common  as  in  England,  being  most  abundant  in  winter, 
when  the  residents  are  outnumbered  by  the  migrants.  They 
appear  on  the  Rock  as  early  as  the  12th  of  August,  but  I  did 
not  detect  them  remaining  to  nest,  although  quantities  breed 
in  the  Cork-wood  and  all  the  wooded  valleys  of  the  neighbour- 
hood, usually  laying  about  the  end  of  April ;  but  in  1894  I 
found  a  nest  touching  the  ground  on  the  Sierra  Retin,  which 
contained  young  about  a  week  old,  on  the  25th  of  March. 

Adult  male.  Forehead  black  ;  head  slate-grey ;  back  rufous  ;  ear- 
coverts  and  underparts  vinous  red  ;  rump  olive-green.  Two  white  bars  on 
wing. 

Male  in  winter.  Duller,  the  slate-grey  feathers  having  brown  tips,  and 
the  white  on  wings  is  tinged  with  yellow. 

Female.  Head  and  back  ashy  brown  ;  below  pale  brown  ;  rump  yellowish 
green. — In  winter  is  browner,  and  the  white  on  the  wings  has  yellow  tinge. 
Length  6  inche.«. 


109.  Fringllla  spodiog-enys,  Bonaparte.     The  North- African 
Chaffinch. 

Moorish.  Birdon,  Bird61. 

This  Chaffinch  is  common  and  resident  in  Morocco.  The 
nest  and  eggs  resemble  those  of  our  Common  Chaffinch,  but  the 
harsher  note  is  very  distinct. 

Mr.  Meade-Waldo  found  them  fairly  common  in  evergreen 
scrub,  and  says  that  they  frequented  more  wooded  ground 
than  the  last  species,  in  the  spring  of  1892.  There  is  no  record 
of  their  occurrence  on  the  Spanish  side  of  the  Straits. 

The  adult  male  is  to  be  distinguished  from  F.  coelebs  by  the  back  being 
yellowish  green,  and  by  the  throat  and  breast  being  light  yellowish  buff. 
Length  G  inches. 

The  females  are  similar  to  those  of  the  Common  Chaffinch. 


104  FRINGILLID^E. 

110.  Fringilla  montifringilla,  Linnams.    The  Brambling. 

Spanish.  Montanes,  Millero. 

This  species  "has  been  once  obtained  near  Tangier,  in  1845, 
when  I  killed  a  female  from  among  a  lot  of  Linnets  ;  the  male 
escaped." — Favier. 

The  Brambling,  on  the  Spanish  side  of  the  Straits,  is  of  very 
irregular  though  not  unfrequent  appearance  near  Gibraltar  in 
winter;  during  that  of  1870-71  they,  as  well  as  Siskins,  were 
abundant.  Near  Seville  they  are  more  regular  in  appearance, 
and  are  sufficiently  well  known  about  Cordova  to  bear  the  local 
name  of  "  Millero." 

Rump  white,  intermixed  with  black  ;  axillaries  and  smaller  inner  wing- 
coverts  pale  yellow. 

Male  in  summer.  Head  and  upper  back  blue-black  ;  throat  and  chest 
orange  ;  flanks  spotted  with  black  ;  bill  bluish. 

Male  in  winter.  Black  parts  edged  with  buff ;  throat  duller,  shaded  with 
sandy  buff  ;  bill  yellow. 

Female.  Head  and  upper  back  dark  brown  ;  rest  of  plumage  much  as 
male  in  winter,  but  less  bright.  Length  6|-  inches. 

111.  Montifringilla  nivalis  (Linnaeus).    The  Snow-Finch. 

This  Alpine  Finch  is  found  in  the  Sierra  Nevada.  Mr.  Stark, 
in  letter,  says  :  "  This  bird  is  very  common  in  Sierra  Nevada,  at 
from  3000  to  6000  feet,  in  small  flocks." 

Head  and  nape  ashy  grey.  Back  brown,  centre  of  feathers  darker  ; 
wing-coverts  and  axillaries  white.  Below  creamy  white.  Black  patch  on 
throat  in  summer.  Bill  black  in  summer,  yellow  in  winter.  Two  centre 
tail-feathers  blackish,  the  rest  white,  all  but  the  outer  tipped  with  black. 
Length  7  inches. 

112.  Linota  cannabina  (Linnaeus).    The  Common  Linnet. 

Moorish.  Sharif  (Favier}.     Spanish.  Camacho,  Jamas. 
"  Abundant  around  Tangier,  many  being  resident  and  nesting 
from  March  to  June.     They  are  mostly  migratory  and  cross  to 


LINOTA  FLAVIROSTRIS.  105 

Europe  in  March  and  April,  returning  in  large  flocks  during 
September  and  October." — Fader. 

The  Common  Linnet  is  very  plentiful  on  the  Spanish  side, 
especially  during  the  winter  months.  Great  numbers  remain  to 
breed,  nesting  in  April,  mostly  in  scrub  on  the  sides  of  the  hills. 
Upon  one  occasion  a  pair  built  on  an  olive-bush  in  my  garden  at 
Gibraltar. 

The  adult  males  are,  as  a  rule,  far  more  brightly  coloured  than 
English  specimens. 

Adult  male.  General  colour  above  reddish  brown  ;  forehead  and  breast 
rich  crimson  ;  rump  paler  than  back  ;  wing-coverts  uniform  chestnut-brown, 
like  the  back,  without  white  edging  ;  throat  whitish.  In  winter  the  red 
colour  is  obscured  by  pale  edgings  to  breast-feathers,  which  wear  off. 

Female.  Has  no  red  on  head  or  breast ;  sides  of  face  dark  ashy,  like  the 
head  ;  bill  dusky. 

Young.  Like  female,  but  spotted.     Length  5^-6  inches. 

113.  Linota  rufescens  (Vieillot).    The  Lesser  Redpole. 

This  bird  is  not  mentioned  by  Favier,  but  is  included  by 
Mr.  Drake  in  his  list  of  the  birds  of  Morocco. 

On  the  Spanish  side  of  the  Straits  it  can  only  be  considered  a 
very  rare  and  irregular  winter  visitant  like  the  Twite. 

Adult  male.  Above  light  brown,  streaked  with  blackish  brown  ;  fore  crown 
carmine  ;  chin  black  ;  throat  and  breast  pink  ;  rump  dusky  brown,  tinged 
with  crimson  and  streaked  with  brown  ;  wing-coverts  tipped  with  rufous 
buff,  forming  a  double  wing-bar. 

Female.  Like  male,  but  only  red  on  the  head.  In  winter  the  pink  colour 
is  much  hidden  by  pale  edges  to  the  feathers. 

Young  dull  coloured,  without  any  red  or  pink.     Length  4£  inches. 

114.  Linota  flavirostris  (Linngeus).    The  Twite  or  Mountain- 

Linnet. 

I  have  no  record  of  the  occurrence  of  this  Linnet  on  the 
Moorish  side.  In  Andalucia  it  is  a  rare  winter  straggler. 

No  red  on  forehead  or  breast  ;  bill  yellow  in  adults,  dusky  in  young. 
Adult  male.  Above  ruddy  brown,  feathers  streaked  with  black  down  the 


106  FKJNG1LLID.E. 

centre,  those  on  the  back  with  pale  margins  ;  rump  purplish  red  ;  throat 
and  fore  neck  clear  reddish  broicn,  the  latter  streaked  with  dark  brown  ; 
axillaries  reddish  brown,  with  black  centres. 

Female.  Like  male,  but  lighter  and  without  the  rosy  rump,  which  is 
uniform  with  the  back.  Length  5-5£  inches. 

115.  Pyrrhula  erythrina,  Pallas.    Scarlet  Grosbeak. 

Three  immature  specimens  of  this  Eastern  Finch  have  been 
obtained  in  October,  near  Malaga. 

The  females  and  young  males  of  this  species  are  at  first  sight  very  likely 
to  be  mistaken  for  immature  Greenfinches  (C.  cMoris},  but  are  to  be 
distinguished  by  the  form  of  the  bill  and  the  two  distinct  yellowish-white 
bars  on  the  wing.  In  this  plumage  it  is  doubtless  the  Fringilla  incerta  of 
Eisso. 

Wings  shorter  than  tail. 

Adult  male.  General  colour  rosy  or  crimson  ;  head  crimson  ;  rump 
brighter  than  back. 

Female.  General  colour  above  olive-brown  ;  feathers  with  dusky  centres ; 
rump  same  colour  as  back  ;  middle  and  greater  wing-coverts  tipped  with 
yellowish  white,  forming  a  double  wing-bar  ;  throat  dull  white,  streaked 
with  brown.  Length  5^  inches. 

116.  Erythrospiza    githaginea    (Lichtenstein).       The    Desert 

Rosy  Bullfinch. 

This  species  is  mentioned  by  Mr.  Drake  as  being  seen  in  the 
south  of  Morocco,  and  has  occurred  accidentally  near  Malaga. 
Several  were  caught  in  1877. 

Adult.  General  colour  rosy.  Female  duller  than  male.  Young  sandy 
brown.  Length  5  inches. 

117.  Loxia  curvirostra,  Linnaeus.    The  Common  Crossbill. 

Spanish.  Pico  cruzado,  Pico  tuerto. 

Favier  only  mentions  having  obtained  this  conifer-frequenting 
bird  once  near  Tangier,  "  a  specimen  being  picked  up  in  a  dying 
state  in  1855." 

Although  never  having  myself  met  with  the  Crossbill  on  either 
side  of  the  Straits,  I  have  seen  undoubted  Andalucian  specimens. 


EMBER1ZA  MILIARIA.  107 

Points  of  bill  crossing  each  other. 

Adult  male.  General  colour  red,  brightest  on  rump  ;  but  sometimes 
breeds  in  yellow  plumage  of  female,  very  faintly  streaked  with  red  in  a  few 
places. 

Female.  Olive-yellow  where  red  in  male,  brightest  on  rump. 

Young.  Like  female,  but  streaked  above  and  below  with  dark  brown. 
Length  (3  inches. 

The  so-called  Parrot  Crossbill  (L.  pityopsittacus)  is  a  Northern  form  of 
the  Common  Crossbill,  with  the  bill  very  much  stouter,  and  is  usually  a 
larger  bird. 

118.  Emberiza  miliaria,  Linnaeus.    The  Corn-Bunting. 

Moorish.  Dorrais  (Favier}.     Spanish.  Triguero ;  Ave  tonta. 

This  well-known  bird  is  exceedingly  numerous  on  both  sides  of 
the  Straits,  and,  being  to  a  great  extent  migratory,  is  perhaps 
least  plentiful  in  winter.  From  their  fearless  stupidity  and 
conspicuous  habit  of  perching  on  the  top  of  some  small  bush  or 
plant,  vast  quantities  are  killed  by  the  Spaniards  and  exposed  for 
sale  in  the  markets,  being  excellent  food,  while  during  the  winter 
months  a  great  many  are  caught  roosting  on  the  ground  by  the 
aid  of  the  lantern. 

The  Common  Bunting  commences  to  lay  about  the  first  week 
in  May,  often  placing  the  nest  at  the  edge  of  marshes ;  and  I 
have  taken  it  in  the  midst  of  a  swamp,  placed  on  a  dry  tussock, 
within  a  yard  of  a  nest  of  Savi's  Warbler.  Verner  remarks  that 
eggs  of  this  Bunting  "  appear  to  undergo  extraordinary  variations 
in  S.  Spain.  I  have  found  many  about  the  vega  of  La  Janda, 
totally  unlike  any  I  have  ever  seen  in  England,  but  of  the 
authenticity  of  which  I  had  no  doubt." 

There  is  a  country  fable  to  the  effect  that  the  Hoopoe  on  first 
arrival  in  Spain  had  no  place  to  go  to,  and  bought  or  hired  some 
land  from  the  Bunting ;  but  when  the  time  for  payment  arrived, 
"Poo,  poo,"  said  the  Hoopoe,  and  flew  away  repeating  his  cry. 
Ever  since  then  the  Bunting  is  continually  and  monotonously 
complaining  of  the  Hoopoe,  and  in  vain  asking  for  his  money. 


108  FEINGILLID^E. 

Sexes  alike.  General  colour  greyish  brown,  streaked  with  dark  brown  ; 
light  eyebrow ;  tail  brown,  with  edges  and  tips  of  buffy  white  ;  no  other 
colours  in  plumage.  Length  7  inches. 


119.  Emberiza  citrinella,  Linnseus.    The  Yellow  Hammer. 
Our  well-known  "Yellow  Hammer,"  so  common  in  the  north  of 

Spain,  is  only  found  in  Andalucia  as  a  winter  straggler.  I  have 
seen  specimens  from  Granada  and  Malaga,  but  never  observed 
any  near  Gibraltar. 

Male  in  summer.  Head  lemon-yellow  ;  back  rufous,  streaked  with  dark 
brown  ;  rump  chestnut ;  under  surface  yellow. 

In  winter.  Head  marked  with  brown,  and  generally  less  bright. 

Female.  Head  less  yellow,  crown  streaked  with  black  ;  throat  and  breast 
striped  with  blackish.  Length  6£-7  inches. 

120.  Emberiza  cirlus,  Linnseus.    The  Girl  Bunting. 

Spanish.  Linacero. 

According  to  Favier's  notes,  this  Bunting  is  equally  common 
around  Tangier  as  near  Gibraltar,  migrating  northwards  in  March 
returning  in  October  and  November,  remaining  during  the  winter 
months ;  many  also  stay  to  breed  during  April. 

The  Girl  Bunting  is  very  frequent  all  along  the  road  from 
Algeciraz  to  Tarifa,  and  from  its  tameness,  and  the  bright  colour 
of  the  males,  with  their  conspicuous  black  throats,  is  sure  to 
attract  attention.  They  appear  to  be  more  a  tree-frequenting 
bird  than  the  Yellow  Hammer  (Emberiza  citrinella),  and  are 
especially  fond  of  the  glades  and  open  patches  of  cultivation  in 
the  Cork-wood. 

Least  wing-coverts  olive-green  ;  rump  olive-brown. 

Male.  Chin,  throat,  and  line  from  bill  through  eye  to  nape  black  ;  eye- 
brow yellow  ;  yellow  crescent  under  the  black  throat ;  upper  chest  dull 
olive,  bordered  below  by  chestnut  ;  belly  dull  yellow. 

Female.  Much  like  female  of  Yellow  Hammer,  but  wanting  the  yellow 
colouring. 

In  icinter,  colours  of  male  less  bright,  the  black  parts  obscured  by  light 
margins.  Length  6  inches. 


EMBEEIZA  SCIKEiVlCLUS.  109 

121.  Emberiza  hortulana,  Linna3us.    The  Ortolan. 

Moorish.  Merskezan  (Favier).     Spanish.  Hortolano. 

''The  Ortolan  is,  next  to  the  Common  Bunting,  the  most 
abundant  of  the  genus  near  Tangier.  Some  remain  to  breed ; 
while  the  rest  pass  on  during  April,  returning  in  September. 
Migrating  in  large  flocks,  they  prefer  wet  ground,  and  are  not 
observed  in  the  winter  months." — Favier. 

Curiously  enough,  I  never  succeeded  in  obtaining  the  Ortolan 
nearer  to  Gibraltar  than  the  vicinity  of  Casas  Viejas,  where  I  found 
them  in  May.  In  the  vicinity  of  Seville  they  are  plentifully  met 
with,  but  do  not  occur  during  the  winter. 

Male.  General  colour  above  pale  reddish  brown  ;  head  greenish  olive  ; 
cheeks,  throat,  and  axillaries  pale  yellow  ;  neck  and  chest  greenish  olive  ; 
rest  below  cinnamon  ;  bill  reddish  brown. 

Female.  Paler  than  male  :  lower  throat  streaked  with  brown.     Length 

7  O 

6  inches. 

122.  Emberiza  schoeniclus,  Linnaeus.    The  Reed-Bunting. 

Included  by  Favier  in  his  list  as  "  rare  near  Tangier ;  met  with 
in  December."  Capt.  Savile  Reid  found  them  common  in  winter 
at  Meshree  el  Haddar.  They  are  most  abundant  near  Gibraltar 
from  December  to  February,  and  were  seen  on  passage  as  late  as 
the  7th  of  April,  They  do  not  remain  to  nest  in  the  sotos  at 
Casas  Viejas  ;  but  near  Seville,  where  they  are  often  sold  in  cages 
under  the  name  of  "  Hortolano,"  I  have  seen  them  in  May,  and 
have  no  doubt  that  they  there  remain  throughout  the  breeding- 
season. 

Male  in  spring.  Head  and  throat  black  ;  cheeks  white  ;  eye-stripe  white ; 
white  collar  from  bill  round  nape  ;  rump  bluish  grey  ;  lesser  wing-coverts 
chestnut. 

In  autumn  the  black  is  hidden  by  rufous  edgings,  and  the  white  by  sandy- 
brown  edgings. 

Female.  Head  and  ear-coverts  reddish  brown,  streaked  with  black  ; 
breast  and  flanks  white,  streaked  with  blackish.  Outside  web  of  second 
tail-feather  white.  Length  6  inches. 


110  FR1NGILL1D.E. 

123.  Emberiza  pusilla,  Pallas.    The  Little  Bunting. 

Nine  specimens  of  this  north-eastern  Bunting  were  obtained 
near  Malaga  on  the  28th  of  December,  1874. 

Male  in  spring.  Centre  of  crown  vinous   chestnut,  with  black  band  on 
each  side  ;  eye-stripe,  lores,  sides  of  face,  and  throat  vinous  chestnut. 
Female.  Like  male,  but  not  so  richly  coloured  on  throat. 
Young.  Throat  white,  no  chestnut  on  breast  or  flanks.    Length  4f  inches. 

124.  Emberiza  cia,  Linnaeus.    The  Rock-Bunting. 

Spanish.  Escribano. 

Although  not  mentioned  by  Favier  as  found  in  Morocco,  I  have 
seen  specimens  obtained  there,  and  met  with  a  pair  in  April  near 
Jebel  Musa. 

On  the  Spanish  side  of  the  Straits  is  a  common  and,  like  most 
of  the  Buntings,  a  stupidly  tame  bird,  as  far  as  my  experience 
goes,  living  about  stony,  rocky,  and  hilly  ground.  Till  1874  I 
never  noticed  them  perching  on  trees  ;  but  in  the  spring  of  that 
year  saw  three  different  birds,  when  disturbed,  settle  on  trees  and 
bushes.  At  Gibraltar  they  are  met  with  in  winter,  but  disappear 
in  the  spring.  I  have  shot  them  at  the  back  of  the  Rock  when 
looking  for  Alpine  Accentors,  in  company  with  which  birds  have 
seen  them  feeding  on  the  refuse-heap  at  the  signal-station.  In 
April  they  frequent  the  slopes  and  tops  of  the  sierras,  nesting 
during  that  month. 

Male.  Centre  of  crown  bluish  grey,  sides  of  crown  blackish.  White 
eyebrow  :  black  stripe  through  eye ;  ear-coverts  bluish  grey  ;  black 
moustachial  stripe  from  bill  to  round  ear-coverts.  Middle  and  greater 
wing-coverts  tipped  with  white,  forming  distinct  wing-bars. 

Female.  Paler.     Length  6  inches. 

125.  Emberiza  saharae,  Tristram.    The  House-Bunting. 

This  bird  is  found  in  the  city  of  Morocco  and  other  Moorish 
towns. 

Back  nearly  uniform  rufous  ;  breast  and  axillaries  sandy  rufous ;  head 
grey,  \vith  blackish  streaks ;  throat  and  fore  neck  light  grey  ;  ashy-grey 
streak  through  eye  and  upper  edge  of  cheek.  Length  5  inches. 


MOTACILLID.fi.  Ill 

126.  Plectrophenax  nivalis  (Linnaeus) .    The  Snow-Bunting. 

The  Snow-Bunting  has  only  been  recorded  once  from  the 
Moorish  side  of  the  Straits  ;  and  this  occurrence  was  mentioned  by 
Mr.  Drake.  This  specimen,  lately  in  the  possession  of  Olcesc 
(Favier's  successor  at  Tangier),  was  a  female,  and  in  fine  plumage. 

In  Andalucia  is  a  very  rare  straggler  in  winter ;  a  male,  which 
I  have  seen,  was  shot  near  Malaga  on  the  1 8th  of  November, 
1872. 

Tail :  six  centre  feathers  black,  fringed  externally  and  round  tip  with 
white  ;  three  outer  pairs  white,  with  black  at  end  of  outer  web  ;  claw  of 
hind  toe  elongated,  but  not  longer  than  hind  toe. 

Male  in  summer.  Head,  neck,  and  under  surface  whits  ;  above  black  ; 
wing-coverts  white  ;  bill  black. 

Female.  Dark  brown  and  white  ;  secondaries  white. 

Male  and  female  in  winter.  Crown  reddish  brown  ;  feathers  on  upper 
parts  broadly  edged  with  reddish  brown  ;  bill  yellowish,  with  dark  tip. 
Length  6^-7  inches. 


MOTACILLIDJE. 
127.  Motacilla  alba,  Linnaeus.    The  White  Wagtail. 

Moorish.  Emzizi.     Spanish.  Pispita. 

"This  is  the  most  abundant  of  the  Wagtails,  and  passes  the 
winter  near  Tangier,  arriving  during  September  and  October, 
leaving  in  March.  They  are  to  be  seen  in  large  flocks  following 
the  plough,  twittering  incessantly." — Fader. 

The  above  notes  apply  equally  to  the  White  Wagtail  in  Anda- 
lucia ;  but  it  was  never  seen  by  me  after  the  24th  of  March ; 
though  Mr.  Stark  found  a  nest  at  Alora,  near  Malaga,  and  they 
are  recorded  by  Arevalo  as  breeding  near  Granada.  They  roost 
together  in  great  numbers  on  the  short  rushes  and  grass  in  the 
marshes  at  Casas  Viejas,  where  I  have  often  seen  them  on  the 
backs  of  horses  and  cattle,  picking  off  vermin  and  catching  flies, 
doing  the  work  of  Starlings  and  Buff-backed  Herons. 


112  MOTACILLID^E. 

Male.  Crown,  nape,  throat;  and  breast  black ;  back  from  nape  ash-grey ; 
front  and  sides  of  head  white. 

Female.  Like  the  male. 

In  winter.  Resembles  the  next  (M.  lugubris),  but  has  lighter  back. 

Young.  Undistinguishable  from  the  young  of  the  next.  Length  7-7^ 
inches. 

128.  Motacilla  lugubris,  Temminck.    The  Pied  Wagtail. 

"  This  is  the  most  rare  of  the  Wagtails  near  Tangier ;  they  are 
found  at  the  same  times  and  places  as  M.  alba" — Favier. 

I  obtained  a  specimen  in  summer  dress  at  Tangier ;  it  is  scarce 
enough  in  Andalucia  in  that  plumage,  but  probably  in  winter 
dress  escapes  unnoticed. 

Male  in  spring.  Crown,  nape,  back,  chin,  and  throat  black,  the  latter 
uniting  with  the  black  on  the  back  and  nape  ;  forehead  and  sides  of  the 
head  white. 

Female.  Like  male,  but  has  the  back  slaty  grey,  with  blackish  feathers 
here  and  there. 

Both  sexes  in  winter  have  the  back  ash-grey  ;  chin  and  throat  white  ; 
back  of  the  head  and  nape  black,  with  a  black  crescent  on  front  of  neck. 

Young.  In  first  plumage  have  entire  head  ash-grey  ;  face  washed  with 
yellow ;  the  black  crescent  on  front  neck  gradually  developing  from  a 
small  spot  to  a  full  crescent.  Length  1-1\  inches. 

129.  Motacilla  melanope,  Pallas.    The  Grey  Wagtail. 

This  Wagtail  is  stated  by  Favier  to  be  a  common  winter  visitor 
near  Tangier,  appearing  in  September  and  October,  departing  in 
February  and  March. 

In  Andalucia  they  are  most  abundant  on  passage  and  during 
the  winter  months,  but  many  pairs  nest  along  the  mountain- 
streams  of  the  sierras — three  or  four  pair  particularly  on  the 
rivulet  which  runs  down  the  Garganta  del  Aguila,  the  valley  in 
which  is  situated  the  "  waterfall "  of  Algeciraz ;  and  they  build 
usually  in  holes  of  the  brickwork  of  the  water-mills,  sometimes 
close  to  the  wheel.  A  pair  also  regularly  nest  at  the  mill  in  the 
Cork-wood ;  and  Mr.  Stark,  when  with  me  on  the  9th  of  April, 
found  a  nest  built  in  a  hole  of  a  large  rock  overhanging  the 


MOTACILLA  FLAVA.  113 

Guadalmalcil,  a  mountain-torrent  between  Tarifa  and  Algeciraz. 
This  nest  was  placed  out  of  reach  of  either  man  or  beast ;  the  hen 
bird  was  visible  from  the  opposite  side,  and  apparently  sitting 
close.  On  the  19th  of  April,  the  nest  at  the  mill  in  the  Cork- 
wood contained  one  egg,  while  another  nest  found  on  the  24th  of 
May  had  four  fresh  eggs  in  it,  which  tends  to  show  that  they 
breed  more  than  once  in  the  season.  The  nest  is  constructed  of 
grass  and  small  roots  lined  with  hair. 

Some  only  of  the  males  had  the  black  throat  on  the  8th  of 
March ;  but  all  had  assumed  it  by  the  1st  of  April.  The  females 
do  not  always  exhibit  this  mark,  some  not  having  it  at  all,  and  in 
none  is  it  so  well  defined  as  in  the  males. 

It  is  the  longest  of  the  Wagtails  found  in  these  parts,  with  the  tail- 
feathers  much  longer  in  proportion  to  the  body  than  in  other  species. 

Under  tail-coverts  brighter  yellow  than  belly. 

Male  in  summer.  Above  bluish  grey  ;  white  eye-stripe  ;  rump  and  upper 
tail-coverts  greenish  yellow  ;  underparts  yellow  ;  throat  black. 

Female.  Resembles  the  male,  but  has  the  throat  white,  or  white  with  very 
rarely  a  few  black  feathers. 

Young.  Above  tinged  with  brown  ;  throat  and  eyebrow  buff. 

In  winter.  Both  sexes  are  alike,  having  a  white  throat.  Length  7-7^ 
inches,  tail  about  3^. 

130.  Motacilla  flava,   Linnseus.     The    Blue-headed  Yellow 
Wagtail. 

Spanish.  Nevadilla. 

This  species  is  found  on  both  sides  of  the  Straits  in  great 
abundance;  the  earliest  seen  was  on  the  20th  and  24th  of 
February  (in  different  years),  many  appearing  on  the  25th.  From 
that  time  till  the  20th  of  April  they  continued  to  pass  ;  and  on 
that  day  I  saw  great  numbers  at  Gibraltar,  resting  on  the  "  flats  " 
at  Europa  after  their  flight  across  the  sea.  They  leave  in  August 
and  September.  Exactly  resembling  our  English  Yellow  Wagtail 
(M.  rail)  in  habits,  they  keep  to  marshes,  nesting  in  grass  and 
herbage  at  the  edge  of  water,  sometimes  among  the  sedges,  and 
lay  in  the  end  of  April. 

I 


114  MOTACILLID^E. 

Male.  Above  yellowish  green  ;  underparts,  except  white  chin,  bright 
yellow  ;  broad  white  stripe  over  eye  from  bill  to  nape  ;  head,  nape,  and  ear- 
coverts  Hue-grey. 

Female.  Head  brownish  and  duller  in  colour.     Length  6-6^  inches. 


131.  Motacilla  borealis,  Sundevall.    The  Grey-headed  Yellow 

Wagtail. 

Occurs  on  passage ;  passing  later  than  the  preceding,  of  which 
it  and  the  next  can  only  be  considered  races. 

Like  the  last,  but  head  slaty  grey  and  has  no  eye-stripe,  and  only  the 
chin  is  white. 

The  race  Motacilla  cinereocapiUa,  Savi,  has  throat  white  and  partly 
defined  white  eye-stripe.  I  shot  specimens  in  spring,  and  they  were 
nesting  near  Casas  Viejas  in  May. 

132.  Motacilla  rail  (Bonaparte).    Yellow  Wagtail. 

This,  the  English  race  of  Yellow  Wagtail,  occurs  on  passage  in 
spring,  but  not  commonly,  and  has  been  obtained  at  Tangier  and 
Malaga.  We  saw  many  about  12th  April,  1894,  at  Tapatanilla. 

Male.  Head  canary-yellow  ;  upper  parts  olive-green  ;  nape  yellowish 
green  ;  all  underparts,  eye-stripe,  and  axillaries  bright  yellow. 

Female.  Eye-stripe  buff ;  plumage  much  less  brilliant. 

Young.  Eye-stripe  and  underparts  huffish  white,  with  no  yellow  except 
a  tinge  on  lower  belly.  Length  6^  inches. 

133.  Anthus  trivialis  (Linnaeus).    The  Tree-Pipit. 

According  to  Favier  this  Pipit  is  common  near  Tangier  during 
migration,  crossing  the  Straits  in  March  and  April,  returning  in 
October  and  November.  On  the  Spanish  side  they  are  found  on 
passage  only :  the  9th  of  April  is  the  earliest  date  on  which  I 
noticed  them,  but  I  saw  many  about  the  20th. 

Above  clear  brown,  distinctly  streaked  with  dusky  ;  spots  on  breast 
small  ;  flanks  streaked  with  black,  like  the  breast  ;  middle  and  greater 
wing-coverts  tipped  with  white,  forming  a  double  wing-bar  ;  light  part  of 
outer  tail-feather  white  or  smoky  white  ;  axillaries  sandy  buff.  Hind  claw 
curved,  shorter  than  hind  toe.  Length  6i  inches. 

/  O  ^ 


ANTHUS  OBSCURUS.  115 

134.  Anthus  pratensis  (Linnaeus).    The  Meadow-Pipit. 

The  Meadow-Pipit  is  equally  common  in  Morocco  and 
Andalucia  from  October  to  the  end  of  March. 

Head  and  upper  parts  olive-brown,  mottled  with  black  centres  ;  rump 
uniform  ;  throat,  chest,  and  flanks  streaked  with  black  ;  light  pattern  of 
tail-feathers  white  ;  axillaries  edged  with  olive-yellow.  Hind  claw  nearly 
straight,  longer  than  hind  toe.  Length  5^  inches. 

135.  Anthus  cervinus  (Pallas).    The  Red- throated  Pipit. 

I  obtained  two  Pipits  in  1870  on  the  10th  of  March,  which 
I  took  home  to  England ;  and  they  were  identified  as  belonging 
to  this  species  by  Mr.  Sharpe  and  Captain  Shelley.  In  1874,  on 
the  8th  of  March,  I  shot  among  a  lot  of  A.  pratensis  another 
bird,  which  appeared  to  be  A.  cervinus;  but  as  the  rufous  throat 
is  not  developed,  though  it  showed  signs  of  that  mark  of  the 
breeding-plumage,  to  determine  with  certainty  to  which  species 
it  really  did  belong  is  impossible ;  so  I  mention  and  include  the 
species  in  my  list  with  a  view  to  some  future  collector  paying 
attention  to  the  subject.  For  my  own  part  I  have  little  doubt 
that  A.  cervinus  does  occur  on  passage  in  Andalucia  and  Morocco. 

Male  in  spring.  Much  resembles  last,  but  has  eye-stripe,  cheeks,  throat, 
and  breast  deep  vinous  red. 

Female.  Onlv  the  throat  vinous. 

if 

In  winter  the  difference  between  the  two  species  is  hardly  distinguishable  ; 
but  it  is  said  that  in  this  species  the  rump  is  spotted  like  the  back,  and  the 
under  tail-coverts  have  dark  centres.  Length  5^  inches. 

136.  Anthus  obscurus  (Latham).    The  Rock-Pipit. 

Favier  states  that  this  species  occurs  in  winter  near  Tangier, 
and  may  be  always  seen  on  the  sea-shore.  I  think  (not  having 
seen  any  of  his  specimens)  that  these  remarks  apply  to  A.  spipo- 
letta,  which  he  does  not  mention  in  his  MS. ;  but  Mr.  Dresser, 
in  his  '  Birds  of  Europe,'  examined  a  specimen  obtained  by 
Olcese  at  Tangier,  and  I  brought  home  a  Pipit  shot  on  the 

i2 


116  MOTACILLID^E. 

mud  at  Palmones,  near  Algeciraz,  in  March  1870,   which  was 
identified  by  Mr.  Sharpe  as  A.  obscurus. 

Axillaries  and  light  pattern  on  outer  tail-feathers  smoky  brown.  Hind 
claw  much  curved,  and  equal  in  length  to  hind  toe. 

Males  in  spring  have  a  rosy  tinge  on  breast,  chiefly  southern  birds. 
Length  6^  inches. 

137.  Anthus  spipoletta  (Linnseus).    The  Water-Pipit. 
Found  on  mud-flats  near  Algeciraz  and  near  Malaga  in  winter. 

White  eye-stripe  ;  below  pale  rosy ;  chest  uniform,  without  streaks. 
Light  pattern  of  outer  tail-feathers  white.  Hind  claw  long  and  straightened. 

Young.  Undistinguishable  in  plumage  from  those  of  A.  obscurus.  Length 
6£  inches. 

138.  Anthus  campestris  (LinnaBus).    The  Tawny  Pipit. 

Moorish.  Solist  (Favier). 

"  Found  near  Tangier  on  passage  in  April,  returning  in 
August,  but  is  not  very  common.  They  migrate  in  pairs  and 
keep  close  together,  so  that  it  is  very  easy  to  shoot  both  at  one 
shot." — Favier. 

The  earliest  date  on  which  I  saw  one  was  at  Tangier  on  the 
31st  of  March.  On  the  Spanish  side  they  appear  to  frequent 
high  ground,  as  on  the  1st  of  May  we  saw  many  on  the  open 
spaces  about  Ojen,  and  thence  all  along  to  the  Venta  de 
Subalbarro.  We  never  met  with  them  on  low  ground,  and 
there  is  no  doubt  they  breed  high  up  on  the  sierras. 

General  colour  of  upper  surface  dull  greyish  brown,  darker  on  centre 
of  feathers  ;  broad  buffy-white  eyebrow  ;  throat  whitish  ;  underparts 
uniform  pale  buff ;  wing-coverts  edged  with  pale  buff,  varying  much  in 
depth  of  colour  ;  light  pattern  of  outer  tail-feathers  white.  Hind  claw 
stout  and  slightly  curved,  about  equal  in  length  to  hind  toe.  Length  6^-7 
inches. 

139.  Anthus  richardi,  Vieillot.    Richard's  Pipit. 

I  shot  one  specimen  of  this  large  eastern  Pipit  on  the  1st  of 
March  on  the  shore,  evidently  just  arrived,  and  obtained  others 


ALAUDID^E.  117 

on  the  20th  of  April,  1870,  not  noticing  any  again;  but  Verner 
reports  that  he  shot  one  on  the  9th  June,  1877,  at  the  stream 
between  the  Pine-woods,  and  Arevalo  records  them  on  passage 
near  Malaga. 

Above  dark  brown,  with  sandy-buff  edges  to  feathers  ;  throat  and  breast 
streaked  ;  axillaries  sandy  rufous  ;  light  pattern  of  outer  tail-feathers  white. 
Hind  claw  long  and  nearly  straight.  Leg  about  1^-  inches  long.  Length 
7 1  inches. 

ALAUDIDJE. 

140.  Certhilauda  bifasciata   (Lichtenstein).      Desert  Curved- 

billed  Lark. 

This  Desert-Lark  is  recorded  from  Malaga  by  Arevalo. 

Hind  toe  very  short.  Bill  long  and  curved.  General  colour  above  and 
centre  tail-feathers  isabelline.  Below  creamy  white  ;  thin  collar  of  blackish 
spots  on  breast ;  eyebrows  white.  Length  8  inches. 

141.  Certhilauda  duponti  (Vieillot).    Small  Curved-billed  or 

Dupont's  Lark. 

Specimens  of  this  Algerian  Lark  were  obtained  near  Malaga 
by  Francisco  de  los  Bios,  and  others  since  his  death. 

Those  found  in  Portugal  have  been  described  as  a  distinct 
species,  C.  lusitanica,  Bocage. 

Bill  long,  slightly  curved,  and  slender.  First  primary  very  short. 
General  colour  sandy  brown,  with  "  lark-like  "  back,  dark  spots  on  back 
with  a  vinous  sheen.  Length  6^  inches. 

142.  Alauda  arvensis,  Linnseus.    The  Skylark. 

Spanish.  Zurriaga,  Terrera. 

"  Found  near  Tangier  during  winter,  arriving  in  October  and 
November,  departing  in  March.  They  are  found  in  immense 
flocks  during  their  stay." — Favier. 

The  same  is  to  be  said  of  the  Skylark  in  Andalucia,  where  in 
some  localities  quantities  are  caught  at  night  with  a  bell  and  a 
lantern,  and  I  have  known  a  boy  bring  in  six  or  seven  dozen  birds 


118  ALAUDID.E. 

at  a  time.     Calandras,   Buntings,  Larks,  in  fact  any  birds  that 
sleep  on  the  ground  can  be  thus  taken. 

Breast  spotted.  Outer  tail-feathers  white,  except  inner  edge  of  inner 
web  ;  next  pair  white  only  on  outer  web.  Hind  claw  long  and  straight. 
Bastard  primary  very  small.  Length  7^  inches. 

143.  Alauda  arborea,  Linnaeus..    The  Woodlark. 

Spanish.  Alondra  de  monte  (Lark  of  the  scrub). 

According  to  Favier,  this  species  "  occurs  near  Tangier  on 
passage  during  March."  Some  were  obtained  by  Capt.  Savile 
Reid  near  Larache  in  the  winter  of  1884-85. 

On  the  Andalucian  side  the  Woodlark  is  sparingly  and  locally 
distributed  during  the  winter  months  up  to  as  late  as  the  21st  of 
April,  frequenting  scrub  where  not  very  thick,  a  favourite  locality 
near  Gibraltar  being  the  chaparales  *  in  the  Cork-wood.  Well 
known  to  the  Spanish  birdcatchers,  and  highly  valued  as  a  cage- 
bird  ;  they  assured  me  that  the  Woodlark  never  remains  to  nest 
near  Gibraltar,  but  they  are  known  to  breed  near  Malaga. 

Light  streak  over  eye  ;  primary-coverts  tipped  with  white,  showing 
small  white  spot  on  closed  wing.  First  primary  long,  third  primary  longest. 
Tail  short ;  outer  feather  greyish  white  at  end,  and  outer  web  bordered 
with  white,  next  three  tipped  with  triangular  white  spots  ;  hind  claw 
curved.  Length  6  inches. 

144.  Alauda  cristata,  Linnams.    The  Crested  Lark. 

Moorish.  Kubaa  (hooded  one — as  hood  of  jellabia,  or  crest  of 
a  helmet).  Spanish.  Carretera. 

The  Crested  Lark  is  one  of  the  most  abundant  birds  both  in 
Morocco  and  Andalucia,  though  never  seen  in  any  great  numbers 
together.  They  are  distributed  in  pairs  on  every  road-track  and 
open  plain,  often  at  intervals  of  only  some  twenty  yards. 
Excessively  tame  and  fearless ,  they  have  acquired  the  name  of 
Carretera,  from  their  habit  of  frequenting  roads,  to  which  they 

*  "  Chaparal,"  lit.  a  plantation  of  ilex.  The  "  chaparales "  here  mentioned  are 
simply  ground  covered  with  brushwood  or  scrub. 


ALAUDA  CRIST  ATA.  119 

resort  as  much  on  account  of  the  horse-  and  mule-dung,  at  which 
they  are  to  be  seen  pecking,  as  for  the  purpose  of  dusting  them- 
selves ;  and  they  are  often  to  be  noticed  on  the  sea-shore,  running 
about  like  a  Sanderling  within  a  yard  of  the  water.  They  have 
no  song  worthy  of  the  name,  and  are  altogether  rather  vulgar 
and  uninteresting  birds.  This  species  is  one  of  those  which  I 
could  not  detect  migrating  in  the  slightest  degree. 

The  Crested  Lark  usually  commences  to  lay  about  the  20th  of 
April,  placing  the  nest  in  some  tuft  of  grass  or  under  shelter  of  a 
small  stone  or  clod  of  earth — constructing  it,  like  those  of  other 
Larks,  with  bits  of  grass,  bents,  &c.,  lined  with  hair. 

One  nest  which  we  found  was  placed  between  the  tracks  of  a 
much  frequented  road  near  Tangier,  in  such  a  position  that  every 
passing  animal  must  have  touched  the  small  clump  of  grass  under 
cover  of  which  the  nest  was  built.  Now,  was  this  site  chosen 
because  snakes,  lizards,  and  other  vermin  would  be  less  likely  to 
come  on  the  beaten  track  I  I  cannot  help  thinking  that  birds  in 
many  instances  have  instinct  enough  to  breed  close  to  houses 
and  roads  with  a  view  to  obtain  protection  from  some  of  their 
enemies  through  the  presence  of  man,  who  is  perhaps  their  worst 
foe. 

The  Crested  Lark  is  subject  to  great  difference  in  the  tints  of 
the  plumage.  The  difference  of  colour  varies  according  to  the 
soil  and  climate,  and  has  been  very  puzzling  to  cabinet  or  table 
naturalists,  as  the  enormous  list  of  synonyms  will  testify. 
Mr.  Dresser,  in  his  article  on  this  bird,  gives  no  less  than  thirty- 
seven  different  names,  which  have  for  the  greater  part  been 
manufactured  on  account  of  the  variations  of  plumage  and  size 
noticed  in  this  species. 

Mr.  Sharpe  (Catalogue  of  Birds,  British  Museum,  vol.  xiii.  p.  633)  has 
Galerita  (Alauda)  thecklce  as  a  distinct  species,  giving  S.  Spain  and  N.E. 
Africa  as  the  habitat.  Exactly  similar  in  habits  and  note,  it  appears  to  be 
only  a  small  local  race  of  A.  cristata. 

Galerita  macrorhyncha,  Tristram,  mentioned  by  Mr.  Drake  as  occurring 
near  Morocco,  is  only  another  local  race  of  the  Crested  Lark. 


120  ALAUDIDJE. 

General  ground-colour  varies  much  from  greyish  brown  to  huffish  brown. 
Conspicuous  occipital  crest  ;  axillaries  buff ;  outer  web  of  outer  tail-feathers 
Intff.  Length  6f  inches. 

145.  Alauda  brachydactyla,  Leisler.    The  Short-toed  Lark. 

Spanish.  Terrera. 

"  This  bird  is  found  on  passage  near  Tangier,  crossing  in 
very  large  flights  during  March  and  April,  returning  in  August 
and  September.  Many  remain  to  breed,  frequenting  the  same 
localities  as  the  Calandra  Lark." — Favier. 

On  the  Andalucian  side  of  the  Straits  the  spring  arrival 
commences  about  the  middle  of  March ;  and  the  passage  con- 
tinues for  a  month  later,  at  which  time  nests  with  eggs  may  be 
found  near  Gibraltar.  Excessively  abundant,  as  above  stated,  in 
the  same  situations  as  the  Calandra  ;  they  prefer  fallow  ground, 
nesting  under  shelter  of  some  clod  or  in  any  slight  depression  of 
the  ground.  I  never  could  find  the  nest,  except  by  putting  the 
old  bird  off.  A  very  good  way  of  finding  the  nests  of  all  the 
Larks  and  ground-breeding  birds  is,  with  the  assistance  of  a  man 
to  hold  it  at  the  other  end,  to  drag  a  rope  about  a  hundred  yards 
long  across  the  ground,  being  careful  that  it  drags  on  the  ground ; 
directly  a  bird  flies  up,  drop  the  rope,  go  to  the  spot,  and  in  all 
probability  a  nest  will  be  found. 

Below  huffish  white  ;  no  spots  on  underparts  ;  but  sometimes  a  streak  or 
two  and  brownish  patch  on  each  side  of  upper  breast.  Outer  tail-feathers 
huffish  white,  except  the  brown  basal  half  of  inner  web  ;  the  next  feather 
has  outer  web  with  buffish-white  edging.  Front  claws  very  short  and 
curved  ;  hind  claw  straight  and  rather  elongated.  Length  5^  inches. 

146.  Alauda  baetica,    Dresser.     The  Andalucian  Short-toed 

Lark. 

Spanish.  Cujailla. 

This  small  Lark  was  discovered  by  Lord  Lilford  in  the  corn- 
land  on  the  banks  of  the  Guadalquivir  below  Seville,  where  they 
are  known  to  occur  from  February  to  the  summer.  Major 


OTOCORYS  BILOPHA.  121 

Verner  found  them  about  the  vega  of  La  Janda,  and  obtained 
several  nests  there  ;  the  eggs  were  much  like  those  of  the  Crested 
Lark.  They  also  occur  in  the  vicinity  of  Malaga  throughout  the 
year. 

Marked  much  as  A.  Irachydactyla,  but  the  secondaries  very  much  shorter 
than  the  primaries.  More  ashy  in  colour  ;  throat,  breast,  and  flanks  very 
broadly  marked  with  black.  Length  5^  inches. 

147.  Alauda  calandra  (Linnaeus).    The  Calandra  Lark. 

Spanish.  Alondra,  Calandria. 

The  Calandra  is  extremely  abundant  and  resident  on  both  sides 
of  the  Straits,  gathering  together  in  flocks  during  the  winter. 

Frequenting  all  the  vegas  or  plains  in  Andalucia,  they  are, 
from  their  numbers,  size,  note,  and  peculiar  varying  flight,  very 
conspicuous,  and  in  some  localities  positively  swarm.  At  times 
their  flight  is  very  like  that  of  some  of  the  smaller  Waders ;  and 
often  when  flying  in  the  breeding-season  they  utter  notes  which 
very  much  resemble  the  cry  of  the  Green  Sandpiper,  only  of 
course  not  so  loud.  In  spring  they  soar  to  a  great  height, 
singing  on  the  wing,  but  not  hovering  like  the  Skylark.  At 
this  season,  when  the  garbanzos  (Cicer  arietinuni)  are  sprouting, 
they  are  very  troublesome,  picking  the  shoots  off,  so  that  the 
farmers  have  to  scare  these  and  other  Larks  away. 

They  consort  much  with  the  Short-toed  Lark,  and  nest  some- 
times close  together  on  cultivated  as  well  as  pasture  land,  laying 
about  the  second  week  in  April. 

The  Calandra  is  a  very  common  cage-bird  at  Gibraltar,  and 
they  are  as  much  prized  for  their  song  as  for  their  lively  habits. 

Bill  short  and  stout,  arched  above.  Large  triangular  black  patch  on 
each  side  of  throat  extending  to  the  neck  forming  an  interrupted  collar. 
Outer  tail-feather  almost  entirely  white.  Length  8  inches. 

148.  Otocorys  bilopha  (Temminck).    The  Horned  Desert-Lark. 
This    species   is   mentioned   by   Mr.    Drake    as   having   been 


122  CY  PSEUDS. 

obtained  at  Morocco ;  and  has  been  found  as  a  straggler  near 
Malaga. 

Hind  claw  long  and  nearly  straight.  Back  pale  fawn-colour,  the  adult 
male  having  a  black  gorget  and  two  long  black  ear-tufts.  Length 
b'  inches. 


Order  PICARI-S1.         Family  CYPSELIDJE. 

Tail  of  ten  feathers. 

149.  Cypselus  apus,  Linnaeus.    Common  Swift. 

Moorish.  Tair  abila  (Favier}.     Spanish.  Avion. 

"This  Swift  arrives  at  Tangier,  en  route  for  Europe,  during 
March  and  April ;  vast  numbers  remain  to  nest  here,  and  return 
south  in  September  and  October." — Favier. 

Mr.  Meade-Waldo  saw  a  Swift  near  Tangier  as  early  as  the 
22nd  of  February,  1892. 

The  above  notes  equally  apply  to  the  Common  Swift  in 
Andalucia,  the  earliest  date  of  arrival  noted  being  the  4th  of 
March,  the  main  body  passing  during  the  last  fortnight  in  that 
month,  some  as  late  as  the  24th  of  April.  The  majority  leave  by 
the  end  of  August,  some  staying  on  into  the  middle  of  September, 
the  last  being  seen  on  the  16th  of  October.  The  number  in 
some  towns,  particularly  Algeciraz,  is  perfectly  marvellous,  and 
the  noise  they  make  morning  and  evening  quite  annoying. 

Verner  writes  : — "  In  order  to  compare  nests  of  this  Swift  with 
those  of  the  next  species,  I  visited  great  numbers  of  nests  of  the 
Common  Swift  on  the  Rock.  In  the  Casemate  Barracks  a  large 
number  nest  yearly  in  the  verandahs,  and  there,  as  elsewhere,  I 
found  all  the  nests,  if  there  were  any,  to  be  only  a  heap  of 
castings  and  insects'  wings,  and  now  and  then  the  remnants  of  a 
Sparrow's  nest.  In  most  cases  the  eggs  were  laid  on  the  bare 
mortar  and  there  was  no  attempt  to  weld  the  castings  &c. 
together  with  mucilage.  I  caught  and  released  a  number  of  old 


CYPSELUS  MURINUS.  123 

birds,  all  of  the  blackest  hue.     Every  nest  contained  two  eggs, 
except  one,  which  held  three. 

General  colour  blackish  brown,  except  small  whitish  chin-patch  ;  four 
toes  all  directed  forwards.  Length  7^  inches. 

150.  Cypselus  murinus,  A.  and  E.  Brehm.      Mouse-coloured 
Swift. 

"  This  Swift  is  found  near  Tangier  on  passage,  crossing  to 
Europe  in  April  and  May.  Some  remain  to  breed ;  but  it  is  the 
least  common  of  the  family,  being  seen  alone  or  in  pairs  in 
company  with  C.  apus.  I  found  a  pair  in  July  1861,  nesting  in 
company  with  some  House-Martins  (Chelidon  urbica} ;  the  nest 
was  simply  an  old  nest  of  that  Martin,  which  the  Swifts  had 
appropriated,  and  contained  two  eggs  of  the  usual  Cypselus 
shape." — Favier. 

They  are  abundant  at  Gibraltar,  and  are  said  to  arrive  at 
Tangier  somewhat  earlier  than  the  Common  Swift,  though  there 
appears  to  be  no  difference  in  the  time  of  their  arrival.  Easily 
noticed  on  the  wing  by  their  light  colour,  they  mix  both  with  the 
Common  and  White-bellied  Swifts. 

In  May  1874,  when  near  Vejer  with  two  ornithologizing 
friends,  we  found  this  species  to  be  more  abundant  than  C.  apus, 
while,  curiously  enough,  at  Algeciraz  (where,  as  mentioned,  there 
are  countless  swarms  of  Common  Swifts)  we  never  could  detect 
one  single  C.  murinus.  They  are  common  around  Malaga ;  and 
Mena,  in  1882,  noticed  their  first  arrival  on  the  14th  of  March. 
He  also  told  me  that  this  Swift  does  not,  like  C.  apus,  enter  the 
nest  direct,  but  settles  or  hangs  outside  before  entering. 

Verner  writes  : — "  In  spite  of  this  Swift  having  been  considered 
only  to  be  a  subspecies,  I  am  convinced  that  in  many  cases  the 
Mouse-coloured  Swifts  nest  in  separate  colonies  apart  from  the 
common  species,  and,  as  far  as  my  experience  goes,  they  differ 
considerably  in  their  nidification. 

"  A  small  colony  used  to  breed  in  the  weep-holes  in  the  sea- 


124  CYPSELID.E. 

wall  of  the  New  Mole  Parade.  Many  black  Swifts  nested  in  close 
vicinity,  but  from  continued  observation  it  appeared  that  the 
Light-coloured  Swifts  nested  in  adjacent  holes  and  kept  apart 
from  their  dark  congeners.  These  holes  being  far  in,  it  was 
difficult  to  be  certain,  since  nests  and  old  birds  were  out  of  reach. 
However,  another  colony  nested  in  1878  in  the  Patio  of  the 
Convent,  and  I  was  able  to  watch  them  daily  and  hourly  for  some 
weeks  from  the  A.D.C.'s  room  which  I  occupied.  I  soon  found 
that  there  were  not  any  dark  Swifts,  and  after  making  sure  of 
this  I  inspected  their  nests.  On  the  12th  June  I  found  five  nests 
— one  with  two  eggs,  three  with  three  eggs,  and  one  with  two 
young  birds  nearly  fledged  ;  some  of  the  eggs  were  fresh,  others 
much  incubated.  The  nests  (placed  on  the  tops  of  rafters  and  on 
a  wall,  and  two  between  a  rafter  and  the  wall)  were  solidly  built 
of  castings  of  insects'  wings,  thistledown,  and  bits  of  paja,  glued 
together  by  mucus  so  strongly  as  to  require  much  force  to  separate 
them  from  the  rafters  :  those  on  the  tops  of  the  rafters  were  much 
smaller  than  the  others,  and  little  more  than  a  saucer-shaped 
rampart.  In  all  cases  I  caught  the  old  birds  and  examined  them 
carefully  before  releasing  them.  It  is  curious  that  the  black 
Swift  usually  lays  two  eggs,  rarely  three,  whilst  three  out  of  four 
of  the  pale  Swifts'  nests  contained  three  eggs." 

151.  Cypselus  melba,  Linnaeus.    White-bellied  Swift,  Alpine 
Swift. 

Moorish.  El  namera.  Spanish.  Avion  real,  Avion  de  pecho 
bianco. 

"  Found  near  Tangier  on  passage,  crossing  the  Straits  from 
March  to  May,  returning  from  August  to  October.  It  is  not  so 
common  as  C,  apus" — Favier. 

The  White-bellied  Swift  breeds  sparingly  at  Gibraltar  in  the 
inaccessible  crevices  of  the  rocks  on  the  Mediterranean  side ; 
they  seem  to  arrive,  if  anything,  a  little  later  than  the  Common 
Swift.  The  earliest  dates  of  arrival  noticed  were  the  24th  of  March 


CAPKIMULQID^.  125 

1871  and  24th  of  March  1894,  Mr.  Irby,  60th  Rifles,  having 
shot  one  at  Tangier  on  the  18th.  On  the  4th  of  April,  1871, 
near  San  Roque,  I  noticed  a  flock  of  about  two  hundred  passing 
in  a  northerly  direction,  with  a  gyrating  flight,  making  a  great 
noise,  though  they  were  very  high  up.  A  few  Common  Swifts 
(their  cry  attracted  my  attention  to  them)  were  with  the  flock. 
On  the  5th  of  November,  1871,  at  Casas  Viejas,  I  saw  six  hawking 
about  over  the  marshes  for  about  an  hour,  when  they  disappeared 
in  a  westerly  direction.  An  oificer  who  was  at  Fez  told  me  that 
he  saw  a  great  many  Alpine  Swifts  there  in  large  flights  about 
the  23rd  of  February. 

Although  far  out  of  the  range  of  this  book,  it  is  interesting  to 
record  that  Major  Verner,  when  at  Crete,  noticed  numbers  of  this 
Swift  passing  north  on  the  24th  of  March,  1886. 

The  sexes  are  alike  in  plumage,  except  that  the  female  is  marked  or 
striated  with  a  faint  black  line  on  the  centre  of  the  feathers  of  the  white 
breast.  The  brain  of  this,  as  in  all  Swifts  that  I  have  examined,  is  small 
for  the  size  of  the  bird. 

Above  greyish  brown ;  chin  and  belly  white,  with  band  of  greyish  brown 
across  upper  breast ;  toes  as  in  last.  Length  8|  inches. 


CAPRIMULGIDJE. 

Mouth  very  wide,  extending  behind  the  eyes ;   tail  of  ten 
feathers  ;  claw  of  middle  toe  serrated  on  the  inner  edge. 

152.  Caprimulgus  europeeus,  Linnaeus.     Nightjar  or  "Goat- 
sucker." 

"  Occurs  near  Tangier,  but  is  less  numerous  than  the  Rufous- 
naped  Nightjar,  some,  however,  remaining  to  nest.  The  others 
pass  on  across  the  Straits  during  May  and  June,  returning  from 
September  to  November  to  pass  the  winter  further  south." — 
Favier. 

The  Nightjar  is  found  in  Andalucia,  as  near  Tangier,  the  earliest 
date  of  arrival  noticed  being  the  5th  of  May. 


126 

Male.  White  spot  near  end  of  inner  web  of  three  first  primaries  of  each 
wing,  and  the  two  outer  tail-feathers  on  each  side  tipped  with  white. 

Female.  Like  male,  but  without  the  white  marks  on  wings  and  tail. 
Length  10£  inches. 


153.  Caprimulg'us  ruflcollis,  Temminck.    Rufous-naped  Night- 
jar. 

Moorish.  Terref  el  rayan.  Spanish.  Zumaya,  Chota  Cabras, 
Engana-pastores  . 

"  This  Nightjar  is  very  abundant  near  Tangier,  arriving  to  cross 
the  Straits  in  April  and  May,  to  return  in  October  and  November. 
Many  remain  to  breed,  nesting  on  the  bare  ground  among  scrubby 
brushwood,  and  laying  two  eggs,  which  are  to  be  found  from 
May  to  August."  —  Favier. 

The  Rufous-naped  Nightjar  is  extremely  plentiful  near 
Gibraltar.  I  knew  one  instance  of  its  occurrence  on  the  16th  of 
February  near  Malaga  ;  the  earliest  date  of  arrival  near  Gibraltar 
was  on  the  15th  of  April,  the  latest  date  of  departure  the  5th  of 
October. 

Much  resembles  the  last,  but  larger,  with  rufous  nape  and  more  ruddy 
chin. 

Sexes  alike,  both  male  and  female  having  white  spots  on  wings  and  tail. 
Length  nearly  12  inches. 


P I C I D  & .     Two  toes  in  front,  two  behind. 

Subfamily  PICIN^E.     Tail-feathers  stiff  and  pointed  : 
nostrils  covered  with  bristles. 

154.  Dendrocopus    major    (Linnseus).      The    Great    Spotted 
Woodpecker. 

This  Woodpecker  is  very  local,  and  is  always  to  be  seen  or 
heard  among  the  old  alder  trees  in  the  Soto  gordo  of  the  Cork- 
wood of  Almoraima.  They  are  found  all  over  that  wood,  up  the 
valleys  (gargantas)  of  the  sierras,  particularly  along  that  one 


DENDROCOPUS  MEDIUS.  127 

down  which  runs  the  river  Palm  ones.  They  are  common  near 
Ojen,  also  abundant  about  Pulverilla  on  the  road  between  Casas 
Viejas  and  Gibraltar.  Further  than  this  I  never  noticed  them ; 
nor  did  I  ever  see  them  where  oak  and  alder  trees  were  absent. 
The  local  name  is  Pito  real.  They  nest  about  the  1st  of  May  in 
holes  of  decaying  trees,  and  do  not  appear  to  be  in  the  slightest 
degree  migratory. 

Black  and  white  ;  scapulars  white  ;   lack  and  rump  black :  under  tail- 
coverts  scarlet ;  iris  red. 
Male.  Nape  scarlet. 
Female.  No  red  on  head  and  nape. 
Young.   Crown  of  head  red.     Length  9|  inches. 

155.  Dendrocopus  numidicus,  Malherbe.    The  Algerian  Pied 
Woodpecker. 

Moorish.  Nakab. 

"  Resident  and  common  in  the  vicinity  of  Tangier,  being  found 
only  in  large  woods,  wThere  they  nest  in  holes  of  trees,  laying 
from  five  to  six  eggs,  similar  to  those  of  Picus  major," — Favier. 

1  did  not  find  this  bird  "  common  "  near  Tangier ;  and  as  for 
the  "  large  woods,"  there  are  none  close  to  that  town ;  about 
Tetuan  this  Woodpecker  is  plentiful,  similar  in  habits  to  I),  major. 
Favier  states  that  they  migrate  across  the  Straits. 

Mr.  Hargitt,  in  his  '  Catalogue  of  Woodpeckers  in  the  British  Museum/ 
vol.  xviii.,  gives  the  Pied  Woodpecker  of  Morocco  as  a  distinct  species, 
D.  mauritamis,  Brehm,  resembling  the  Algerian  D.  numidicus,  except  that 
the  black  stripes  which  border  the  throat  and  extend  to  sides  of  the  chest 
do  not  form  a  complete  band  tipped  with  scarlet  on  the  chest,  as  in 
D.  numidicus  ;  further  that  the  lateral  tail-feathers  are  more  broadly 
marked  with  black,  and  the  bill  and  legs  less  powerful.  Andalucian 
specimens  of  D.  major  often  have  traces  of  crimson  and  black  crescent  on 
the  chest ;  and  until  a  larger  series  of  specimens  are  examined,  I  think 
numidicus  and  mauritanus  are  only  climatic  races  of  D.  major. 

[Dendrocopus  medius.    The  Middle  Spotted  Woodpecker. 
This  Pied  Woodpecker  is  stated  by  Arevalo  to  be  resident  in 


128  PICID,E. 

the  Province  of  Granada.     At  Potes,  near  Santander,  in  1876, 
we  found  this  species  at  a  higher  elevation  than  D.  major. 

Both  sexes  have  the  crown  vermilion,  the  abdomen  and  tail-coverts  rose- 
colour.  Length  8£  inches.] 

156.  Gecinus  sharpii,  Saunders.      The  South  Spanish  Green 

Woodpecker. 

Spanish.  Pito  real. 

This  Green  Woodpecker,  in  habits,  note,  and  manner  of 
nesting,  is  exactly  similar  to  the  British  G.  viridis;  they  are 
abundant  in  some  localities  near  Seville,  particularly  in  the  Cotos 
and  towards  Cordova  and  Granada.  I  never  met  with  any  nearer 
Gibraltar  than  the  vicinity  of  Seville,  and  it  is  very  singular  they 
should  not  occur  in  the  Cork-wood  or  in  the  wooded  gargantas 
or  valleys  in  the  vicinity  of  Gibraltar. 

Resembles  G.  viridis,  but  has  the  side  of  the  face  grey  and  the  red 
moustache  of  the  male  not  bordered  with  black.  Length  11^-12^  inches. 

157.  Gecinus  vaillanti    (Malherbe).      The    Algerian    Green 

Woodpecker. 

Moorish.  Nakab  el  tebak  (the  borer  of  the  wood). 

"  Kesident  near  Tangier,  but  not  so  common  as  Picus  numi- 
dicus ;  like  the  latter  avoiding  the  haunts  of  men  and  living  in 
large  woods.  They  nest  in  holes  of  trees  in  April  and  May,  and 
lay  from  five  to  eight  shining  white  eggs.  The  males  assist  in 
incubation." — Favier. 

We  found  this  Green  Woodpecker  to  be  common  near  Tetuan 
and  in  the  province  of  Angera,  especially  among  the  short  stunted 
trees  which  grow  in  the  valleys  about  Jebel  Musa ;  they  are  rare 
near  Tangier,  the  scarcity  of  trees  accounting  for  their  absence  ; 
in  habits  and  note  they  exactly  agree  with  G.  viridis  and 
G.  sharpii,  but  both  the  latter  and  the  present  bird,  in  my 
humble  opinion,  are  only  climatic  or  local  races  of  G.  viridis. 

The  characters  given  are  : — Face  grey,  with  whitish  stripe  across  from 


IYNX  TORQUILLA.  129 

base  of  upper  bill  to  end  of  moustache,  which  is  black  in  both  sexes  ;  but  I 
have  seen  male  specimens  from  Morocco  with  the  moustache  red. 

Male.  Crown  and  nape  red. 

Female.  Crown  grey,  striped  with  black  ;  nape  only  red.  Length  12 
inches. 

158.  Gecinus    canus,    Gmelin.      The     Grey-headed     Green 
Woodpecker. 

I  have  never  met  with  this  species  in  Andalucia ;  but  there 
was  a  specimen  in  an  Institute  at  Seville,  stated  to  have  been 
obtained  in  the  neighbourhood.  Lord  Lilford  records  them  from 
the  vicinity  of  Madrid,  and  Arevalo  from  Valencia. 

The  note  of  this  Green  Woodpecker  is  quite  distinct  from  that 
of  G.  viridis. 

Nape  and  head  grey.     Below  uniform  ;  moustache  black  "in  both  sexes. 

Male.  Fore  crown  red. 

Female.  Crown  grey.     Length  8  inches. 


Subfamily  lYNGIN^E.     Nostrils  partly  covered  by  a  membrane. 

159.  lynx  torquilla,  Linnseus.    The  Wryneck. 

Spanish.  Torcecuello,  Hormiguero,  Lililo. 

"Rather  scarce  and  seen  only  in  pairs  near  Tangier  during 
passage,  crossing  the  Straits  in  March  and  April,  returning 
in  August  and  September,  but  occasionally  observed  up  to 
December." — Favier. 

Mr.  Meade- Waldo  saw  one  near  Tangier  at  the  end  of  January, 
1892. 

On  the  Spanish  side  of  the  Straits  the  Wryneck  is  occasionally 
heard  in  March,  April,  and  September ;  probably  their  line  of 
migration  lies  further  to  the  east. 

General  colour  greyish  brown,  vermiculated  with  blackish ;  below  dull 
white  ;  throat  and  breast  tinged  with  buff  and  crossed  with  blackish  bars. 
Tail  soft  and  rounded,  with  six  blackish  bars  ;  the  outer  feather  on  each 
side  so  extremely  short  as  to  give  the  appearance  of  only  ten  feathers. 
Length  7  inches. 


130  COKAC11DJE. 


ALCEDINID.3E. 

160.  Alcedo  ispida,  Linnseus.    The  Kingfisher. 

Moorish.  Kandil  el  behar  (lamp  of  the  sea).  Spanish.  Martin 
pescador. 

"  This  bird,  only  found  from  August  to  March,  is  not  numerous 
near  Tangier,  but  is  more  abundant  near  Rabat." — Favier. 

The  Kingfisher  is  common  in  winter  and  spring  near  Gibraltar, 
being  frequently  seen  among  the  rocks  on  the  coast,  and  often 

the  "  inundation  "  at  the  North  Front ;  and  they  have  bred 
near  Gibraltar,  where  the  majority  arrive  in  October,  leaving  in 
March.  Arevalo  says  they  are  resident  in  Spain. 

Male.  Above  greenish  blue  ;  middle  of  back,  rump,  and  upper  tail-coverts 
cobalt-blue  ;  throat  and  patch  on  each  side  of  neck  white  ;  lores  black  ; 
rest  below  rich  chestnut ;  bill  black.  Length  7^  inches. 

Female.  Lower  mandible  reddish  at  base. 


CORACIIDJE. 

161.  Coracias  garrula,  Linnseus.    The  Roller. 

Moorish.  Sharrakrak.     Spanish.  Carlanco,  Carraca. 

"This  bird  is  seen  in  numbers  near  Tangier  on  passage, 
migrating  in  pairs  and  crossing  the  Straits  in  April  and  May, 
returning  in  August  to  retire  further  south.  Their  food  is  all 
kinds  of  insects,  even  scorpions." — Favier. 

The  Roller  breeds  at  Larache,  nesting  in  holes  of  the  walls  of 
the  ramparts  at  the  end  of  April.  I  did  not  observe  any  else- 
where in  Morocco,  except  about  the  ruins  of  "  old  Tangier." 

In  Andalucia  they  are  also  very  local,  and  arrive  during  the 
latter  end  of  March,  leaving  by  September.  I  have  seen  one  or 
two  in  May  near  Casas  Viejas,  one  12th  April,  1894,  Sierra  Retin, 
while  Verner  saw  one  at  Pulverilla,  August  20,  1878  ;  but  they 
are  not  common  nearer  to  Gibraltar  than  the  vicinity  of  Seville. 
Thence  along  the  valley  of  the  Guadalquivir  they  abound,  nesting 


181 

in  holes  of  trees,   walls,  and  ruins ;   they  lay  about  the  14th  of 
April,  from  four  to  six  shining  white  eggs. 

Back,  scapulars,  and  tertials  cinnamon-brown  ;  inner  web  of  end  of 
primaries  blackish  brown.  Neck,  throat,  breast,  and  underparts  bluish  ; 
rest  green  or  purplish  green.  Length  13  inches. 


MEPOPIDJE. 

162.  Merops  apiaster,  Linnaeus.    The  Bee-eater. 

Moorish.  El  Leeamoon.     Spanish.  Abejaruco. 

"  The  Bee-eater  is  seen  on  passage  near  Tangier  in  great 
flights,  which  attract  notice  from  their  cry.  They  arrive  and 
cross  over  to  Europe  during  March  and  April,  returning  in 
August,  many  remaining  to  breed.  They  nest  in  May,  the  eggs 
varying  in  shape,  being  some  oval,  some  oblong." — Favier. 

This  bird  did  not  appear  to  me  to  be  quite  so  common  in 
Morocco  at  the  end  of  April  as  on  the  Spanish  side  of  the 
Straits,  where,  during  April,  May,  June,  and  July,  it  is  one 
of  the  most  conspicuous  birds  in  the  country;  at  that  season 
Andalucia  without  Bee-eaters  \vould  be  like  London  without 
Sparrows.  Everywhere  they  are  to  be  seen ;  and  their  single 
note,  teerrp,  heard  continually  repeated,  magnifies  their  number 
in  imagination.  Occasionally  they  venture  into  the  centre  of 
towns  when  on  passage,  hovering  round  the  orange-trees  and 
flowers  in  some  patio  or  garden.  Crossing  the  Straits  for  the 
most  part  in  the  early  part  of  the  day,  flight  follows  flight  for 
hours  in  successiori,  always  exactly  in  the  same  direction,  due 
north.  When  passing  at  Gibraltar  they  sometimes  skim  low 
down  to  settle  for  a  moment  on  a  bush  or  a  tree,  but  generally 
go  straight  on,  often  almost  out  of  sight ;  but  their  cry  always 
betrays  their  presence  in  the  air. 

My  dates  of  their  first  arrival  noticed  are : — the  7th  of  April, 
1868;  4th  of  April,  1869;  1st  of  April,  1870;  29th  of  March, 


132 

1871;  26th  of  March,  1872;  28th  of  March,  1874;  29th  of 
March,  1876  (Verner);  28th  of  March,  1877  (Verner) -,  29th  of 
March,  1894.  None  were  seen  on  the  30th,  many  on  the  31st  of 
March,  1894 ;  none  on  the  1st,  2nd,  3rd  of  April,  many  on  the 
6th ;  some  on  the  7th,  many  on  8th  and  9th,  most  on  the  10th  ; 
numbers  on  llth,  12th,  13th,  and  14th.  They  were  observed 
passing  in  great  numbers  from  the  10th  to  the  14th  of  April 
in  three  consecutive  years,  the  greatest  quantity  arriving  on  the 
10th;  so,  in  Spanish  fashion,  I  christened  that  date  "St.  Bee- 
eater's  day."  The  latest  flight  I  ever  saw  going  north  was  on  the 
7th  of  May. 

Having  remained  at  Gibraltar  once  only  during  July  and 
August,  I  had  but  that  opportunity  of  watching  the  return 
migration,  which  appeared  during  the  last  week  in  July  and  also 
on  the  llth  and  12th  of  August,  the  last  being  noticed  on  the 
29th  of  that  month,  all  with  few  exceptions  being  heard  passing 
at  night.  Verner  noticed  the  return  migration  at  Gibraltar  on 
the  20th  of  July  and  on  the  18th  of  August,  earliest  and  latest 
dates.  The  first  arrivals,  as  is  the  case  with  all  migrants,  are 
those  which  remain  to  breed  in  the  immediate  neighbourhood. 
Commencing  their  labours  of  excavation  almost  immediately  they 
arrive,  the  earliest  eggs  that  I  know  of  were  taken  on  the  29th 
of  April ;  but  usually  they  do  not  lay  till  about  the  second  week 
in  May,  often  not  so  soon. 

In  some  places  they  nest  in  large  colonies ;  in  others  there  are 
perhaps  only  two  or  three  holes.  When  there  are  no  river-banks 
or  barrancos  in  which  to  bore  holes,  they  tunnel  down  into  the 
ground,  where  the  soil  is  suitable,  in  a  slightly  vertical  direction, 
on  some  elevated  mound. 

The  shafts  to  these  nests  are  not  usually  so  long  as  those  in 
banks  of  rivers,  which  sometimes  reach  to  a  distance  of  eight  or 
nine  feet  in  all ;  the  end  is  enlarged  into  a  round  sort  of  chamber, 
on  the  bare  soil  of  which  the  usual  four  or  five  shining  white 
eggs  are  placed ;  after  a  little  they  become  discoloured  from  the 


MEKOPS  APIASTER.  133 

castings  of  the  old  birds,  the  nest  being,  as  it  were,  lined  with  the 
wings  and  undigested  parts  of  bees  and  wasps.  Vast  quantities 
of  eggs  and  young  must  be  annually  destroyed  by  snakes  and 
lizards :  the  latter  are  often  seen  sunning  themselves  at  the 
entrance  of  a  hole  among  a  colony  of  Bee-eaters  ;  and  frequently 
have  I  avenged  the  birds  by  treating  the  yellow  reptile  to  a  charge 
of  shot.  The  bills  of  Bee-eaters,  after  boring  out  their  habita- 
tions, are  sometimes  worn  away  to  less  than  half  their  usual 
length ;  but  as  newly  arrived  birds  never  have  these  stumpy  bills, 
it  is  evident  that  they  grow  again  to  their  original  length.  It  has 
often  been  a  source  of  wonder  to  me  how  they  have  the  strength 
to  make  these  long  tunnels  ;  the  amount  of  exertion  must  be 
enormous  ;  but  when  one  considers  the  holes  of  the  Sand-Martin, 
it  is  perhaps  not  so  surprising  after  all. 

During  my  stay  at  Gibraltar,  Bee-eaters  decreased  very  much 
in  the  neighbourhood,  being  continually  shot  on  account  of  their 
bright  plumage  to  put  in  ladies'  hats.  Owing  to  this  vile  fashion, 
we  saw  no  less  than  seven  hundred  skins,  all  shot  at  Tangier  ill 
the  spring  of  1874,  which  were  consigned  by  Olcese  to  some 
dealer  in  London.  However,  the  enormous  injury  these  birds  do 
to  the  peasants  who  keep  bees,  fully  merits  any  amount  of 
punishment;  but  at  the  same  time  they  destroy  quantities  of 
wasps.  After  being  fired  at  once  or  twice,  they  become  very 
wary  and  shy  at  the  breeding-places ;  and  the  best  way  to  shoot 
them  is  to  hide  near  the  colmenares,  or  groups  of  cork  bee-hives 
(corchos),  which  in  Spain  are  placed  in  rows  sometimes  to  the 
number  of  seventy  or  eighty  together ;  and  it  is  no  unusual  thing 
to  see  as  many  Bee-eaters  wheeling  round  and  swooping  down, 
even  seizing  the  bees  at  the  very  entrance  of  their  hives. 

The  reason  of  their  early  departure  in  August  is  to  be  accounted 
for  by  the  simple  fact  that  bees  cease  to  work  when  there  are  no 
flowers,  as  by  that  time  all  vegetation  is  scorched  up. 

Bill  long,  pointed,  slightly  curved  downwards;  iris  red.  Forehead  white 
next  bill,  then  a  bluish  tinge  ;  crown  rich  chestnut ;  chin  and  throat  rich 


134  UPUPID^E. 

yellow,  bordered  below  by  a  black  line.     Tail  green,  long  ;  two  centre 
feathers  an  inch  longer  than  others.     Length  10-11  inches. 


UPUPIDJE. 

163.  Upupa  epops,  Linnaeus.    The  Hoopoe. 

Moorish.  Hudhud,  Tair  ababil,  Kubaa  deen  Nasara  (the 
Christian's  Crested  Lark).  Spanish.  Abubilla,  Gallo  de  Marzo, 
Cajonera,  Cagajonera,  Sabubilla. 

"  Seen  in  great  quantities  near  Tangier  on  passage,  crossing  to 
Europe  during  February,  March,  and  April,  returning,  to  retire 
altogether  for  the  winter,  in  August,  September,  and  October. 
In  some  years  the  vernal  migration  is  earlier,  and  they  are  seen 
at  the  end  of  January.  They  rarely  remain  to  nest  near  Tangier. 
The  females  have  a  nearly  white  throat.  The  superstitious  Jews 
and  Mahometans  both  believe  that  the  heart  and  feathers  of  the 
Hoopoe  are  charms  against  the  machinations  of  evil  spirits." — 
Earner. 

'  Hoopoes  seldom  remain  to  nest  in  the  vicinity  of  Gibraltar : 
Verner,  in  1875,  found  a  nest  in  a  hollow  tree  in  the  Sierra 
Retin,  which  contained  four  fresh  eggs  on  31st  May,  and  a  few 
breed  about  Casas  Viejas,  and  thence  northwards,  where  there 
are  trees ;  towards  Moron  and  Seville  their  "  hood,  hood"  may 
be  frequently  heard  in  spring  and  summer.  They  begin  to  lay 
about  the  1st  of  May,  in  holes  of  trees. 

My  dates  of  their  earliest  arrival  at  Gibraltar  are: — the  17th 
of  February,  1870;  18th  of  February,  1871;  IGth  of  February, 
1872;  17th  of  February,  1894;  but  on  the  llth  of  January  in 
1872  I  saw  a  single  Hoopoe  in  the  Goto  del  Eey.  Verner  first 
noticed  them  in  1875  on  17th  of  February;  1876  on  the  18th; 
in  1879  on  the  21st;  Mr.  Irby,  60th  Eifles,  on  17th  of  February, 
1892 ;  whilst  Capt.  Savile  Reid  informed  me  of  one  appearing  as 
early  as  the  16th  of  January  in  1874.  They  mostly  pass  in  March, 
whence  their  local  name  Gallo  de  Mdrzo,  March-cock. 


CUCULID.E.  135 

Bill  long,  thin,  slightly  arched  and  pointed.  Erectile  rich  buff  crest, 
tipped  with  black  ;  the  hind  feathers  of  the  crest  with  white  before  the 
black  tips.  Tail  black,  of  ten  feathers,  with  broad  white  bar  across  the 
lower  part.  Length  10-12  inches. 


CUCULIDJE.     Tail  of  ten  feathers ;   two  toes  in  front, 

two  behind. 

164.  Cuculus  canorus,  Linnaeus.    The  Common  Cuckoo. 

Moorish.  Takouk,  Ukouk.     Spanish.  Cuco. 

"  More  abundant  near  Tangier  than  the  Great  Spotted  Cuckoo ; 
seen  during  passage,  in  pairs,  which  cross  to  Europe  in  April 
and  May,  and  return  in  August  to  winter,  probably,  in  the  inte- 
rior of  Africa.  Some  remain  during  summer,  awaiting  the  return 
of  the  autumnal  migration." — Favier. 

The  Cuckoo  is  very  plentiful  near  Gibraltar,  especially  in  the 
Cork-wood  and  on  all  hill-sides  wherever  there  are  any  trees.  I 
saw  a  great  many  at  the  top  of  the  mountains  at  the  back  of 
Algeciraz  at  the  end  of  May,  but  not  beyond  the  line  of  trees. 
First  heard  on  the  7th  of  April  in  1868,  on  the  22nd  of  March 
in  1870,  on  the  31st  of  March  in  1871,  on  the  29th  of  March  in 
1872,  and  on  the  30th  of  the  same  month  in  1874;  on  the  llth 
of  March,  1879  (Verner);  24th  of  March,  1894.  They  remain 
till  the  end  of  July. 

A  female  shot  in  the  second  week  in  May  had  then  two  eggs 
remaining  in  the  ovaries,  nearly  ready  to  lay.  Verner  found  on 
25th  June,  1879,  near  Gibraltar,  a  Cuckoo's  egg  in  a  Wood- 
Warbler's  nest. 

Male.  Above  bluish  grey ;  chin  and  neck  ash-grey  ;  below  white,  barred 
with  black.  Tail  greyish  black,  graduated,  slightly  spotted  and  tipped  with 
white  ;  iris,  eyelids,  legs,  and  feet  yellow. 

Female.  Like  male,  but  usually  with  slight  rufous  tinge  on  breast. 
Females  much  less  numerous  than  males. 

Young.  Upper  surface  and  tail  clove-brown,  barred  with  pale  brown, 
feathers  tipped  with  white  ;  white  spot  on  nape  ;  iris  brown.  Length 
14  inches. 


136  CUCULID.E, 

165.  Coocystes   grlandarius  (Linnaeus).      Tlie  Great  Spotted 
Cuckoo. 

Moorish.  Teir  el  Keber  (Bird  of  news)  (Favier).  Spanish. 
Cuco  real,  Cuco  monon. 

"  Occurs  near  Tangier  on  passage,  always  in  pairs,  but  not  in 
any  great  numbers.  They  cross  to  Europe  in  January,  February, 
and  March,  returning  in  June,  July,  August,  and  September. 
Their  food  is  entirely  caterpillars,  both  smooth  and  hairy." — 
Favier. 

The  Great  Spotted  Cuckoo  arrives  in  Andalucia  much  earlier 
than  the  Common  Cuckoo ;  and  though  Favier  states  that  they' 
pass  in  January,  the  25th  of  February  and  the  2nd  of  March  are 
the  earliest  dates  which  I  have  for  their  arrival,  and  they  mostly 
appear  between  the  7th  and  28th  of  March.  Verner  records  one 
on  8th  March  and  another  on  4th  April,  1879,  both  near  Gibraltar. 
The  latest  I  saw  was  on  the  7th  of  August,  in  the  Alameda  at 
Gibraltar ;  but  they  are  seldom  noticed  near  there,  and  pass  on 
to  districts  further  north,  where  there  are  Magpies  (Pica  rustica), 
as  they  lay  in  the  nests  of  the  latter,  and  occasionally,  it  is  stated, 
in  those  of  the  Spanish  Magpie  (Cyanopicus  cooJci).  The  eggs 
can  be  easily  distinguished  by  their  elliptical  form,  those  of  the 
Magpie  being  pointed  at  one  end.  They  vary  a  good  deal  in 
size  and  much  in  the  markings,  like  those  of  the  bird  whose  nest 
they  use.  As  far  as  we  yet  know,  this  Cuckoo  always  places  its 
eggs  in  nests  of  the  Corvidce.  The  majority  of  eggs  I  have  seen, 
mostly  obtained  by  Ruiz  of  Seville,  came  from  the  vicinity  of 
Cordova ;  there  are  a  good  many  in  the  Goto  del  Key.  A  female 
killed  on  the  7th  of  March  had  the  eggs  so  far  developed  as  to 
show  that  the  probable  number  of  eggs  she  would  have  laid  was 
four. 

The  Rev.  John  White  mentions  this  Cuckoo  as  having  been 
killed  at  Gibraltar  about  1776. 

Adults.  Crested.    Head  and  nnpe  bluish  grey;  rest  of  upper  parts  greyish 


STRIGID^E.  137 

brown  ;  wings,  scapulars,  and  tail  (except  two  centre  feathers)  tipped  with 
white  ;  underparts  white.     Sexes  alike. 

Young.  Basal  two-thirds  of  primaries  chestnut;   head  blackish   brown, 
without  any  crest.     Length  15-16  inches. 


Order  STRIGES.  Family  STRIGIDJE. 

166.  Strix  flammea,  Linnseus.    The  White  or  Barn-Owl. 

Moorish.  Youka,  Tair  el  m6t  (Death-bird),  Sehar,  Bou  tezaz. 
Spanish.  Lechuza. 

"This  Owl,  resident  near  Tangier,  is  nearly  as  abundant  as 
the  Little  Owl,  inhabiting  ruins  and  holes  in  rocks,  and  nesting 
twice  a  year,  between  April  and  November.  They  lay  from 
three  to  four  eggs.  The  inhabitants  of  Tangier  consider  this  bird 
the  clairvoyant  friend  of  the  Devil.  The  Jews  believe  that 
their  cry  causes  the  death  of  young  children ;  so,  in  order  to 
prevent  this,  they  pour  a  vessel  of  water  out  into  the  courtyard 
every  time  that  they  hear  the  cry  of  one  of  these  Owls  passing 
over  their  house.  The  Arabs  believe  even  more  than  the  Jews ; 
for  they  think  that  they  can  cause  all  kinds  of  evil  to  old  as  well 
as  to  young  ;  but  their  mode  of  action  is  even  more  simple  than 
that  of  their  antagonists  the  Jews,  as  they  rest  contented  with 
cursing  them  whenever  they  hear  their  cry.  Endeavouring  to 
find  out  from  the  Mahometans  what  foundation  there  is  for  the 
evil  reputation  of  this  species,  I  was  told  this : — '  When  these 
birds  cry,  they  are  only  cursing  in  their  language ;  but  their 
malediction  is  harmless  unless  they  know  the  name  of  the 
individual  to  whom  they  wish  evil,  or  unless  they  have  the 
malignity  to  point  out  that  person  when  passing  him;  as  the 
Devil  sleeps  but  little  when  there  is  evil  work  to  be  done,  he 
would  infallibly  execute  the  command  of  his  favourite  if  one  did 
not,  by  cursing  the  Owl  by  name,  thus  guard  against  the  power 
of  that  enemy  who  is  sworn  to  do  evil  to  all  living  beings.' 


138  STRJGID^E. 

Having  learned  the  belief  of  the  Mahometans  relative  to  this 
Owl,  it  was  more  difficult  to  find  out  exactly  that  of  the  Jews, 
who  when  questioned  by  me  knew  not  how  to  answer,  except 
that  the  act  of  pouring  out  water  in  the  middle  of  the  courtyard 
is  a  custom  of  long  standing  in  order  to  avert  the  evil  which  the 
Owl  is  capable  of  doing ;  that  is  to  say,  the  water  is  poured  out 
with  a  view  of  attracting  the  Evil  Spirit's  attention  to  an  object 
which  distracts  him,  and  so  hides  from  him  the  infant  which  the 
Owl  in  its  wickedness  wishes  to  show  him." — Favier. 

On  the  Andalucian  side  of  the  Straits  the  White  Owl  is 
common  and  resident,  nesting  at  Gibraltar  in  the  Moorish 
Castle. 

I  must  here  digress  to  say  a  few  words  in  favour  of  this  most 
useful  of  birds.  Almost  exclusively  feeding  on  rats  and  mice, 
they  deserve  every  encouragement  and  support  that  can  be  afforded 
them  ;  but  from  being  in  all  countries  regarded  with  superstitious 
awe  and  dislike,  they  are  more  or  less  persecuted  on  that  account ; 
and  in  England,  through  the  ignorance  and  stupidity  of  game- 
keepers, who  fancy  that  they  kill  game  (i.  e.  feathered  game),  they 
suffer  most  severely.  This  excuse  is  ridiculous  ;  for  the  old  birds 
they  have  not  the  power  to  kill,  and  young  pheasants  and 
partridges  at  the  time  the  Owls  are  on  the  feed  are  safely  being 
brooded  by  the  parent  bird. 

Those  who  wish  .to  encourage  and  increase  Owls,  and  have 
not  hollow  trees  or  buildings  where  they  nest,  may  always  gratify 
their  wishes  by  fixing  an  empty  barrel  (about  an  18-gallon  size) 
horizontally  in  the  fork  of  any  large  tree,  cutting  a  hole  in  one 
end  large  enough  for  the  birds  to  enter ;  but  the  hoops  of  the 
cask  should  be  screwed  on,  or  it  will  soon  fall  to  pieces.  Not 
only  the  Barn-Owl,  but  the  Tawny  Owl  (Syrnium  aluco)  also  will 
use  these  barrels  or  "  owl-tubs."  The  difficulty,  however,  is  to 
keep  out  Jackdaws  ;  but  when  once  the  Owls  have  established 
themselves,  there  is  no  fear  of  that  intrusion. 

In  a  barrel   put  up  too  near  another  in  which  was  an  Owl's 


ASIO  ACCIPITEINUS.  1  39 

nest,  a  pair  of  Stock  Doves  took  possession  and  reared  their 
young.     This  same  tub  afterwards  had  a  hornet's  nest  in  it. 

Facial  disk  complete  and  large,  tuftless  ;  skin  covering  orifice  of  ear 
(opercaluni)  large. 

Above  pale  orange-buff,  speckled  with  grey  and  spotted  with  black  and 
white  ;  underpants  and  facial  disk  white,  but  these  parts  vary,  the  former 
being  sometimes  marked  with  black  specks,  and  both  are  occasionally  of  a 
complete  buff  colour.  Iris  black.  Toes  without  feathers,  covered  with  a 
few  bristles  ;  middle  claw  serrated.  Length  14  inches. 

167.  Asio  otus  (Linnaeus).    The  Long-eared  Owl. 
Spanish.  Carabo. 

This  arboreal  Owl  is  not  mentioned  by  Favier,  though  Olcese 
obtained  one  near  Tangier  in  1884,  where  Mr.  Meade- Waldo 
found  it  the  most  common  Owl,  except  the  Little  Owl  and  Scops, 
and,  as  in  the  Canaries,  frequenting  rough  scrub-covered  ground. 

I  did  not  meet  with  any  near  Gibraltar,  but  shot  one  in  winter 
in  the  Goto  del  Rey.  They  are  found  near  Cordova  and  Granada 
and  breed  there. 

Facial  disk  complete  ;  ear-tufts  very  long,  equal  to  hind  toe  and  claw  ; 
operculum  semicircular  ;  iris  orange.  General  colour  orange-buff  ;  above 
mottled  with  blackish  brown,  below  streaked  and  faintly  barred  with  brown. 
Length  14  inches. 

168.  Asio  accipitrinus  (Pallas).    The  Short-eared  Owl. 

Moorish.  El  hama  (Favier).     Spanish.  Carabo. 

"This  species  occurs  less  abundantly  than  the  Cape-Owl 
(A.  capensis),  being  found  on  passage  in  small  flights  on  open 
and  wet  ground.  Some  breed  near  Tangier  ;  but  the  remainder 
cross  to  Europe  in  February  and  March,  returning  in  November. 
This  Owl  interbreeds  with  the  Cape-Owl,  producing  hybrids 
which  only  differ  from  that  species  in  having  the  front  of  the 
facial  disk,  the  throat  and  tarsi  whitish,  while  the  irides  are  half 
yellow.  The  Arab  chasseurs  confound  the  two  species  under 
the  name  of  '  el  hama ' ;  but  they  are  easily  distinguished  by  the 


140  STKIGID^E. 

irides,  which  are  yellow  in  the  present  species,  and  hazel  in  the 
Cape-Owl." — Favier. 

The  above  story  about  the  hybrids  is  difficult  to  believe,  and 
is  to  my  mind  apocryphal ;  however,  it  is  given  for  what  it  may 
be  worth.  I  confess  I  am  very  sceptical  as  to  the  assertions 
made  about  the  interbreeding  of  different  species  in  an  absolute 
state  of  nature,  excepting  only  the  Gallinse ;  but  most  hybrids 
among  these  are  produced  under  circumstances  of  acclimatization 
which  can  hardly  be  called  a  really  wild  state. 

The  Short-eared  Owl  may  nest  so  far  south,  but,  as  far  as  my 
observations  go,  is  in  Andalucia  only  a  winter  visitor,  and  even 
then  not  very  abundant.  I  should  have  omitted  this  story  of 
Favier's  but  that  it  has  appeared  in  print  before. 

Disk  and  operculum  like  last ;  ear-tufts  very  short. 
Above  tawny,  each  feather  with  dark  brown  down  the  centre  ;  below 
buff,  streaked  with  blackish  brown  ;  iris  yellow.     Length  14-10^  inches. 

169.  Asio  capensis  (Smith).    Marsh-Owl. 

Moorish.  El  hama. 

"  Is  a  common  resident  near  Tangier,  usually  frequenting  wet 
swampy  ground,  feeding  chiefly  on  insects.  Some  pass  over  to 
Europe  in  March  and  April,  returning  in  November  and 
December.  They  nest  on  the  ground  in  April  or  May,  laying 
four,  rarely  five,  round  white  eggs,  sometimes  marked  with  a  few 
rusty  spots.  The  young  are  not  always  hatched  at  the  same 
time,  as  in  the  same  nest  may  be  found  young  birds  of  diiferent 
growths." — Favier. 

My  experiences  of  this  Owl  in  Spain  are  very  limited,  and  as 
follows: — In  October  1868,  on  my  first  visit  to  Casas  Viejas, 
when  looking  for  Snipe  in  one  of  the  wettest  parts  of  the  Mill 
soto,  two  Owls  rose  at  my  feet,  which  I  shot,  winging  one,  which 
I  carried  home  alive  to  take  to  Gibraltar,  seeing  at  once  from  the 
bluish-black  colour  of  the  irides  that  I  had  got  an  Owl  which  I 
did  not  know.  Afterwards  hunting  about,  one  more  was  seen,  and 


SYRNIUM  ALUCO.  141 

killed.  On  the  10th  of  November  following,  during  my  second 
visit,  I  saw  three  more,  two  of  which  I  winged  and  also  carried 
off  alive  to  Gibraltar,  keeping  them  there  for  some  time,  till  one 
got  out  and  flew  off  as  if  nothing  was  the  matter  with  it ;  so  I 
sent  the  other  at  once  to  Lord  Lilford,  who  had  it  alive  till  1870. 
I  met  with  no  more  till  the  10th  of  November,  1870,  when  I 
shot  one  and  picked  up  the  remains  of  another.  In  October  and 
November  1871  I  repeatedly  and  carefully  went  over  the  same 
ground,  but  did  not  see  any,  while  friends  of  mine  there  in 
August  and  September,  whom  I  begged  to  look  out  for  these 
Owls,  did  not  come  across  one.  All  the  eight  birds  above 
mentioned  were  found  within  a  space  of  about  a  square  mile ; 
and,  strange  to  say,  I  never  saw  any  elsewhere.  In  December 
1873,  Capt.  Savile  Reid,  of  the  Royal  Engineers,  shot  one  when 
snipe-shooting  in  the  same  locality.  I  was  there  in  March  and 
May  in  1874,  and,  though  I  hunted  all  the  likely  ground  over, 
failed  to  meet  with  even  one  ;  but  they  have  since  been  obtained 
there,  notably  in  the  winters  of  1882-83  and  1893-94. 

General  colour  above  earthy  brown,  indistinctly  vermiculated  with 
rufous.  Ear-tufts  barely  visible.  Iris  bluish  black.  Length  13-14  inches. 

170.  Syrnium  aluco  (Linnseus).    The  Tawny,  Wood,  or  Brown 
Owl. 

Moorish.  Lu  Lual,  Bu-ru-ru,     Spanish.  Carabo. 

"  This  species  is  the  scarcest  of  the  Owls  near  Tangier,  being 
met  with  on  passage,  crossing  to  Europe  in  February,  returning 
in  November  and  December.  Some  remain  to  nest  in  April, 
laying  two  eggs,  of  which  often  only  one  is  hatched.  They  live 
in  large  thick  woods." — Favier. 

Mr.  Meade-Waldo  says  the  Tawny  Owls  which  he  saw  in 
Morocco  were  very  grey  birds  and  very  large ;  all  specimens 
which  I  saw  from  that  country  were  of  the  grey  race.  The 
Arabic  name  Bu-ru-ru  is  delightfully  suggestive  of  their  cry. 

I  never  met  with  or  heard  one  in  Andalucia,  and  it  is  not 


142  STRIGID^E. 

mentioned  by  Arevalo  as  an  Andalucian  bird ;  but  Mr.  Stark  saw 
one  in  April,  1876,  near  Algeciraz,  and  another  was  obtained  by 
Mena  in  winter  near  Malaga. 

Facial  disk  complete ;  no  tufts  ;  operculum  large ;  toes  feathered  ; 
iris  blackish.  Two  phases  of  plumage  occur  :  grey  and  rufous.  Length 
18-19  inches. 

•171.  Scops  giu  (Scopoli).    The  Scops  Owl. 

Moorish,  Maroof  ("the  well  known").  Spanish.  Corneja, 
Corneta,  Cuquillo. 

"  Occurs  near  Tangier  on  passage,  crossing  to  Europe  in  March, 
returning  to  winter  further  south  in  September  and  October. 
Many  pass  the  breeding-season  in  Morocco." — Favier. 

The  Scops  Owl  is  very  plentiful,  both  in  Morocco  and  Anda- 
lucia,  but  is  almost  entirely  migratory.  I  was  much  surprised  to 
hear  one  on  the  13th  of  January,  1872,  near  the  Goto  del  Rey, 
and  another  nearer  Seville  on  the  15th  ;  but  from  what  I  after- 
wards was  told  at  Seville,  there  is  no  doubt  a  few  sometimes 
remain  there  during  the  winter  ;  I  never  observed  them  during 
that  season  at  Gibraltar,  the  earliest  date  of  the  vernal  migration 
noticed  being  the  4th  of  March,  the  first  nest  being  on  the  4th 
of  May.  This  Owl  always  nests  in  holes  of  trees,  at  least  I  do 
not  know  of  any  instance  of  their  nesting  in  rocks  or  ruins,  like 
Carine  noctua,  which  breeds  by  preference  in  those  places. 

Abundant  in  the  Cork-wood  ;  the  nest  is  easily  discovered  by 
going  round  and  hammering  at  the  old  cork-trees  with  a  stick, 
when,  if  a  Scops  Owl  flies  out,  ten  to  one  there  is  a  nest.  They 
are  strictly  arboreal,  and  their  monotonous  single  note  may  be 
frequently  heard  repeated  at  regular  intervals  by  day  as  well  as  by 
night ;  they  even  frequent  trees  in  the  midst  of  towns,  being  often 
heard  in  the  trees  which  fringe  the  Delicias,  the  drive  and  Rotten 
Row  of  Seville. 

They  chiefly  feed  on  Coleoptera,  and  I  believe  are  entirely 
insectivorous. 


BUBO  IGNAVUS.  143 

Facial  disk  incomplete  above  eyes,  tufted  ;  no  operculum.  General 
colour  grey,  with  minute  marks  of  brown  and  spots  of  dark  brown.  Legs 
feathered  ;  toes  bare.  Iris  yellow.  Length  8^-10  inches. 


172.  Bubo  ignavus,  T.  Forster.    The  Eagle-Owl. 

Spanish.  Bujo  real. 

This  Owl  is  not  included  in  Favier's  notes  on  the  birds  of 
Tangier,  though  it  is,  no  doubt,  found  in  the  mountainous 
districts  of  Morocco ;  indeed,  we  heard  of  a  large  Owl  about 
Tetuan,  but  could  not  obtain  a  specimen. 

They  are  resident  in  all  rocky  localities  in  Andalucia ;  and 
some  frequent  the  "Rock,"  probably  nesting  in  some  of  the 
numerous  inaccessible  caverns  of  the  east  side.  One  was  caught 
in  1869  in  a  magazine  near  the  Rock  gun;  having  gone  down 
into  the  narrow  space  between  the  outer  and  the  main  wall  of  the 
magazine,  it  was  unable  to  rise.  I  had  this  bird  alive  for  some 
time,  and  ultimately  sent  it  to  Lord  Lilford,  in  whose  possession 
it  paired  with  another  from  Norway.  I  also  had  three  young 
from  a  nest  near  Castellar,  about  eighteen  miles  from  Gibraltar. 
While  keeping  these  Owls,  the  wild  ones  used  to  come  at  night 
close  to  the  cage  and  answer  the  call  of  those  that  were  shut  up. 
Its  loud,  melancholy,  human-sounding  note  is  sometimes  to  be 
heard  all  night  long  up  the  Rock,  and  is  usually  supposed  to  be 
the  cry  of  the  apes. 

They  breed  very  early :  judging  from  the  size  of  the  young 
which  were  obtained,  they  would  lay  about  the  end  of  January; 
and  such,  say  the  cabreros,  is  the  case.  I  never  could  succeed  in 
discovering  the  nest,  though  I  knew  of  several  reputed  nesting- 
places,  but  on  examining  them  found  nothing  but  bones  of 
rabbits,  rats,  partridges,  and  small  birds,  never  even  seeing  one 
of  the  Owls,  though  the  charcoal-burners,  carboneros,  assured 
me  that  they  had  taken  the  young  from  these  situations.  One 
man,  however,  said  that  these  Owls  bring  the  .young  from  their 
nests  to  these  caves.  Verner  reports  that  on  1st  of  April,  1894, 


144  STRIGID.E. 

he  saw  an  Eagle-Owl  fly  round  a  hill  where  he  had  seen  a  pair 
twenty  years  before ;  climbing  up  he  discovered  the  nest,  which 
was  merely  a  basin-shaped  hollow  in  the  dry  sand  under  an  over- 
hanging crag:  round  the  nest  was  a  great  mass  of  bones  and 
castings,  and  half  a  rabbit,  a  water-vole,  and  part  of  a  peewit — 
not  a  common  bird  in  those  parts  in  April.  The  nest,  which 
contained  one  young  bird  and  an  addled  egg,  was  well  concealed 
in  the  thick  tall  heather  and  most  difficult  to  find.  There  was 
a  second  disused  nesting-place  within  twenty  yards. 

The  Eev.  John  White  mentions  the  Eagle-Owl  as  occurring  at 
Gibraltar  during  his  residence  there  about  1776. 

Facial  disk  incomplete  above  eyes  ;  tufts  very  large;  no  operculurn. 

General  colour  above  blackish,  mottled  with  tawny  yellow.  Toes  so 
thickly  feathered  that  the  last  joint  is  hidden.  Iris  orange-yellow.  Length 
26-27  inches. 

173.  Carine  noctua  (Scopoli).    Little  Owl. 

Moorish.  Mouka,  Moueka,  Bouma.     Spanish.  Mochuelo. 

"  Is  the  commonest  species  of  Owl  near  Tangier,  being  both 
resident  and  migratory.  Those  which  migrate  pass  to  Europe 
during  March  and  April,  returning  in  August  and  September. 
During  passage  they  are  met  with  in  pairs  or  small  flights ;  at  all 
other  times  they  are  found  singly  or  in  pairs  among  large  rocks 
and  old  buildings." — Favier. 

The  above  was  written  by  Favier  under  the  head  of  Athene 
ylaux,  or,  as  he  had  it,  "  Strix  noctua  meridionalis" ;  but  as  that 
species  has  not  yet  been  noticed  in  Andalucia,  and  as  C.  noctua 
is  the  Little  Owl  of  Tangier,  I  have  no  hesitation  in  referring 
the  above  notes  to  that  bird — not  that  it  would  have  been  much 
loss  to  have  omitted  them  altogether,  the  only  information  of 
importance  being  that  they  migrate. 

Near  and  at  Gibraltar  this  Little  Owl  is  common  and  resident, 
nesting,  about  the  end  of  April,  in  holes  of  trees  as  well  as  in 
rocks. 


VULTURID.E.  145 

Facial  disk  ill-defined  ;  (uftless  ;  no  operculum. 

Above  greyish  brown,  spotted  with  white  ;  below  white,  much  streaked 
with  brown  ;  iris  yellow  ;  toes  without  feathers,  but  covered  with  bristles. 
Length  7^-8  inches. 

174.  Carine  glaux,  Savigny.    Southern  Little  Owl. 

This  bird,  which  is  only  a  light-coloured  race  of  the  last,  does 
not  appear  to  be  met  with  in  the  immediate  vicinity  of  Tangier, 
the  only  specimens  I  have  seen  having  been  obtained  three  or 
four  days'  journey  on  the  way  to  Fez.  The  Little  Owl  of  Tangier 
is  undoubtedly  C.  nociua. 

Resembles  Little  Owl,  except  is  isabelline  or  pale  fawn-colour  where  dark 
brown  in  that  bird. 


Order  ACCIPITRES.      Family  VULTURID^l. 

175.  Vultur  monachus,  Linnaeus.     The  Black  or  Cinereous 
Vulture. 

Spanish.  Buitre  negro. 

This  Vulture  is  mentioned  by  Favier  as  having  once  occurred 
near  Tangier  ;  and  there  is  a  specimen  in  the  Norwich  Museum 
from  that  locality,  perhaps  the  identical  bird.  They  are  probably 
not  so  rare  in  Morocco  as  Favier  implies.  On  the  Spanish  side 
of  the  Straits  they  are  not  often  to  be  seen,  and  generally  alone 
in  winter  and  early  spring. 

They  are  more  common  near  Seville  than  Gibraltar.  Some 
breed  in  Andalucia,  as  I  discovered  one  nest  by  watching  the 
birds  building  or,  rather,  repairing  it ;  for  on  examination  it 
appeared  to  be  an  old  nest,  probably  a  Stork's,  and  was  a  vast 
pile  of  sticks  placed  on  a  half-decayed  alder  tree,  about  fifteen 
feet  from  the  ground,  in  the  midst  of  the  thick  jungle  of  the  Soto 
Malabrigo,  nearCasas  Viejas.  This  place  is  almost  impenetrable, 
surrounded  by  open  marsh,  and  is  formed  of  a  mass  of  huge 
tussocks  placed  far  apart,  on  which  grow  wide-spreading  sallows 

L 


146  VULTURID.E. 

and  brambles  well  interlaced.  The  space  between  these  tussocks 
is  covered  with  rushes  and  sedges,  growing  in  mud  and  water,  in 
places  up  to  the  waist.  In  my  first  expedition  to  the  nest  it  took 
me  more  than  half  an  hour  to  reach  the  tree,  a  distance  of  only 
about  a  hundred  and  fifty  yards  from  the  edge  of  this  paradise  of 
Water-Rails  and  aquatic  Warblers.  Upon  climbing  the  tree  it 
was  very  difficult  to  see  into  the  nest,  as  it  so  overhung,  owing 
to  the  great  breadth ;  and,  alas !  there  was  no  egg,  not  even  any 
lining. 

A  few  days  after,  on  the  26th  of  February,  I  again  examined 
the  nest,  only  to  find  it  lined  with  wool  and  a  few  dried  rushes. 
Muster-day  at  Gibraltar,  on  the  28th,  compelled  our  return  to  the 
Rock ;  so  I  engaged  a  man  to  take  the  nest  and  bring  the  egg  to 
Gibraltar,  which  he  never  did,  probably  not  liking  the  journey 
through  the  swampy  jungle.  The  following  year  this  nest  was 
not  used  by  any  birds;  but  in  1874  a  pair  of  White-shouldered 
Eagles  took  possession,  repairing  the  nest  and  lining  it  with  fresh 
green  boughs.  This  was  early  in  March ;  and  with  persistent 
bad  luck,  on  our  return  there  in  April,  my  friend  found  nothing 
in  the  nest,  although  the  Eagles  were  about ;  either  they  had 
been  robbed  of  their  eggs,  or  else  had  deserted  owing  to  too 
frequent  examination.  Verner  took  a  nest,  in  the  spring  of  1875, 
of  the  White-shouldered  Eagle  from  this  tree,  which  was  subse- 
quently burnt  down,  and  he  observed  a  pair  of  Black  Vultures 
about  the  vicinity  of  the  Malabrigo  during  that  and  the  four 
following  springs.  He  told  me  that  once  he  came  across  a  large 
flock  of  Griffons  feeding  on  a  dead  horse,  and  among  these  birds 
was  a  single  Black  Vulture,  which  he  watched  and  noticed  was 
master  of  the  situation  and  drove  away  any  Griffon  that  attempted 
to  feed  too  near. 

The  Black  Vulture  is  said  to  nest  near  Utrera;  but  upon 
inquiry  I  could  not  ascertain  such  to  be  the  case ;  they  appear  to 
go  further  north  to  breed,  as  Lord  Lilford  found  them  nesting 
towards  Madrid,  and  in  one  season  received  no  less  than  some 


GYPS  FL'LVUS.  147 

seventy  eggs.  More  solitary  in  habits  than  the  Griffon,  and 
unlike  that  Vulture,  they  build  in  trees  and  not  in  colonies — • 
laying  only  one  egg,  about  the  beginning  of  April. 

This  Vulture  is  to  be  recognized  when  on  the  wing,  within  a 
short  distance,  by  its  dark  appearance.  The  immature  birds  are 
very  dark-coloured,  becoming  lighter  with  age,  till  they  attain 
the  adult  plumage.  The  bare  skin  about  the  head  and  neck  is 
of  a  pale  bluish  colour ;  nostrils  rounded. 

Adult.  Ruff  of  brown  down.     Plumage  much  lighter  than  young. 
Young.  Ruff  of  lanceolated    feathers,  general    colour    blackish   brown. 
Length  42  inches. 

176.  Gyps  fulvus  (J.  F.  Gmelin).    Griffon  Vulture. 

Moorish.  Niser.    Spanish.  B nitre,  Pajaraco,  Buitre  franciscano. 

"  This  Vulture  occurs  commonly  near  Tangier,  both  as  a 
resident  and  on  passage,  and  is  often  seen  feeding  in  company 
with  the  Neophron  on  the  same  carcass." — Foxier, 

I  did  not  see  many  Griffon  Vultures  in  Morocco,  but  there 
wrere  a  few  pair  about  Jebel  Musa  in  April.  Near  Gibraltar  they 
are  very  plentiful,  nesting  in  colonies,  not  exceeding  thirty-five 
pairs,  in  holes  or,  rather,  small  caves  in  the  perpendicular  crags 
or  "  lajas,"  which  are  found  in  many  of  the  Sierras  ;  but  occasion- 
ally a  solitary  nest  will  be  found  some  distance  from  any  other, 
while  Mr.  A.  C.  Stark  mentions  a  nest  "  built  on  the  ground  near 
the  top  of  a  hill "  in  the  Tarifa  district. 

The  most  important  breeding-places  near  to  Gibraltar  are  the 
Sierra  de  San  Bartolome,  the  Sierra  de  Plata,  and  la  Laja  del 
Ciscar,  all  near  Taivilla.  One  egg  only  is  the  usual  comple- 
ment ;  and  they  lay  about  the  20th  of  February.  Should  the 
first  egg  be  taken,  it  seems  that  they  lay  again  about  the  15th  of 
April.  Of  course  it  is  impossible  to  prove  this ;  but  eggs  were 
laid  at  that  time  in  nests  which  had  been  robbed  in  February. 
Stark  says  that  in  the  neighbourhood  of  Gibraltar  and  Tarifa 
fresh  eggs  of  the  Griffon  may  be  obtained,  roughly  speaking, 

L2 


VU1TURID.E. 


NEST  AND  EG3  OF  GRIFFON"  ON  OPEN  LEDGE  200  FEET  FROM 
SUMMIT  OF  EL  CISCAR. 


GYPS  FULVUS.  149 

from  the  1st  of  February  to  the  1st  of  March,  the  majority  laving 
between  the  10th  and  20th  of  February.  Near  Malaga  they 
are  somewhat  earlier,  many  laying  in  the  latter  half  of  January. 
The  egg  is  usually  white,  but  is  occasionally  marked  with  buff- 
coloured  blotches,  the  nest  consisting  sometimes  merely  of  three 
or  four  bits  of  green  bush  laid  on  the  rock,  but  being  generally 
lined  with  green  materials,  mostly  broom,  esparto  grass,  rushes, 
and  one  with  furze  (Stark),  and  the  birds,  like  other  Raptores 
that  use  green  stuff  to  line  their  nests,  continually  renew  it. 

Verner  obtained  "  fifteen  eggs  on  February  the  25th,  1879,  six 
of  which  he  blew,  and  found  two  fresh,  three  slightly,  and  one 
much  incubated.  One  of  the  fifteen  was  faintly  spotted  with 
rufous  at  the  large  end ;  the  others  were  quite  white."  On  15th 
of  March,  1878,  he  took  "  eight  eggs,  one  spotted  with  rufous  at 
the  small  end ;  the  others  were  white :  four  of  the  eggs  were 
fresh,  four  considerably  incubated."  On  the  15th  of  April  he  got 
"  two  eggs,  which  contained  perfectly-formed  embryos,  about  size 
of  newly-hatched  Jackdaws,  in  which  the  shape  of  the  bill,  feet, 
and  claws  of  the  adult  Griffons  was  clearly  discernible.'*  On 
May  14th,  1875,  he  got  "two  almost  quite  fresh  eggs,  no  doubt 
laid  by  birds  whose  first  eggs  had  been  taken.  As  regards  the 
colour  of  eggs :  when  first  laid  they  are  pure  white,  but  soon 
become  stained  and  often  covered  with  mud  and  blood.  Anyone 
who  has  seen  a  party  of  Griffons  on  damp  soil,  churning  up  the 
ground  with  their  feet  around  a  carcass,  can  easily  understand 
the  eggs  becoming  soiled.  They  are  essentially  cavern-haunting 
birds,  and  whenever  possible  construct  their  nests  in  cavernous 
situations — occasionally  in  isolated  crags,  very  rarely  on  an  open 
ledge,  as  shown  in  the  illustration  on  p.  148 ;  but  the  engraving 
of  the  young  Griffon  in  the  nest,  and  that  of  the  cave  (p.  150), 
where  two  pairs  of  birds  were  nesting,  better  exemplify  their 
usual  breeding-places  "  *. 

*  The  Griffon  was  found  nesting  en  an  oak  tree  in  Slavouia  by  the  late  Crown 
Prince  of  Austria. 


150 


VULTURID.fi. 


CAVE  WITH    GRIFFONS'  NESTS. 


YOUNG  GRIFFON  IN  C.VYERN. 


GYPS  FULVUS.  151 

Stark  also  mentions  that  many  castings,  about  four  inches  long 
by  one  inch  in  diameter,  composed  entirely  of  vegetable  matter, 
such  as  long  grass,  fibres,  and  green  leaves  of  the  cork-tree,  are 
to  be  found  near  their  nests  and  roosting-places ;  these  materials 
do  not  appear  to  have  been  "  accidentally  devoured  with  the 
intestines  of  animals,  but  are  probably  plucked  and  swallowed 
by  the  Vultures  to  cleanse  their  stomachs,  for  when  freshly 
ejected  they  are  coated  with  a  tough  glairy  mucus." 

It  is  a  fine  sight  to  see  thirty  or  more  of  these  gigantic  birds 
fly  out  at  once  with  a  rushing  noisy  flight  from  their  nests,  which 
they  do  if  a  shot  is  fired  at  the  bottom  of  the  cliff  in  which  they 
breed  ;  and  this  is  the  only  method  of  finding  the  exact  position 
of  their  nests,  as  well  as  those  of  other  rock-nesting  birds,  though 
later  on  each  large  crevice  or  hole  where  there  is  a  nest  is  plainly 
visible,  owing  to  the  dung  which  covers  the  face  of  the  rock 
below,  looking  as  if  a  bucket  of  whitewash  had  been  poured  out 
of  the  cave.  Vultures  in  Andalucia  are  far  more  wary  than  in 
other  countries  in  which  I  have  seen  them,  except,  of  course, 
during  the  breeding-season. 

Griffons  roost  according  to  the  wind,  generally  about  rocks,  and 
near  their  breeding-places ;  but  we  found  in  the  Sierra  Retin 
many  roosting  on  cork-trees  on  the  side  of  a  sheltered  valley. 
Some  of  these  trees  were  nearly  killed  by  the  Vultures'  droppings, 
and  the  ground  all  around  strewed  with  their  feathers,  showing 
they  must  have  used  this  situation  for  many  years.  Verner  informs 
me  that  on  the  opposite  side  of  the  great  cliffs  where  they  breed 
on  the  Ciscar,  he  has  seen  as  many  as  eight  together  taking 
shelter  in  one  large  sandstone  cavern  on  the  lee  side  of  the 
mountain. 

How  the  numbers  which  inhabit  Andalucia  at  times  find 
sufficient  to  eat  is  a  puzzle  to  me ;  they  must  be  able  to  fast  for 
some  days,  or  else  travel  immense  distances  for  their  food,  as  in 
the  winter  and  spring  it  is  unusual  to  see  dead  animals  about ; 
but  in  the  hot  parching  summer  months  vast  quantities  of  cattle 


152  VULTURIDJS. 

die  of  thirst  and  want  of  pasture.  A  bull-fight  is  a  sort  of 
harvest  to  Vultures,  which  flock  in  great  numbers  to  revel  on  the 
carcases  of  the  unfortunate  horses  that  have  been  so  cruelly 
killed. 

The  Griffon  Vulture  may  be  distinguished  on  the  wing  by  its 
light  colour  when  within  reasonable  distance. 

Nostrils  perpendicular,  slightly  oval.  General  colour  ashy  fulvous  ; 
head  and  neck  covered  with  whitish  down  ;  ruff  of  white  down.  Primaries 
and  tail  blackish  ;  iris  light  yellow. 

Young.  Much  darker ;  ruff  of  tawny  lanceolate  feathers  ;  iris  hazel. 
Length  40  inches. 

177.  Neophron  percnopterus  (Linnaeus).    Egyptian  Vulture. 

Moorish.  Rekhama.  Spanish.  Monigero  (near  Gibraltar), 
Relijero,  Alimocha,  Pernetero,  Abanto,  and,  away  from  the 
Sierras,  Quebrantahuesos. 

"  Appears  near  Tangier  in  flocks  during  migration,  some 
remaining  to  nest  in  the  vicinity,  awaiting  the  return  of  the 
autumnal  migration,  to  winter  probably  in  the  interior  of  Africa. 
Those  which  pass  over  to  Europe  cross  from  February  to  April, 
returning  in  August  and  September.  They  nest  on  rocks  in 
April,  generally  laying  two  eggs,  sometimes  only  one.  These 
have  a  rough  surface,  and  vary  in  shape.  Sometimes  there  is  an 
interval  of  two  or  three  days  in  the  hatching  of  eggs  in  the  same 
nest.  Fifty-four  eggs  have  passed  through  my  hands." — Favier. 

Near  Gibraltar,  Neophrons,  during  their  stay,  are  abundantly 
distributed.  Two  pair  nest  regularly  on  "  the  Rock,"  going  by 
the  name  of  "  Rock-Eagle "  among  those  who  would  call  a 
Buzzard  a  Bustard,  and  vice  versa.  One  nest  is  below  O'Hara's 
Tower,  the  other  below  the  Rock  gun  on  the  North  Front. 

Many  pass  northwards  at  the  end  of  February,  the  23rd  of  that 
month  in  1870,  and  the  same  day  in  1892,  at  Tapatanilla,  being 
the  earliest  dates  on  which  they  were  observed  by  me ;  and  the 
greater  number,  many  hundreds,  almost  always  in  pairs,  pass 
during  March.  On  the  21st  and  24th  of  that  month,  in  1872, 


NEOPHftON  PERCXOPTERUS. 


153 


NEST  OF  NEOPHRON,  SHOWING  CRAG. 


NEST  OF  NEOPHRON,  NEARER  TIEW. 


154  VULTUKID.E. 

great  quantities  crossed  at  the  same  time  as  flights  of  Booted 
Eagles,  Snake-Eagles,  Common  Buzzards,  Ked  and  Black  Kites ; 
Verner  noticed  numbers  passing  at  Gibraltar  on  the  25th  of 
March  1877,  31st  of  March  1878,  and  28th  March  1879. 

The  Neophron  usually  begins  to  lay  about  the  1st  of  April. 

Verner  found  eggs  on  the  6th  of  April,  slightly  incubated,  and 
on  the  13th  of  April  two  quite  fresh  ;  he  says  "  the  second  egg 
is  always  laid  some  days  later  than  the  first  one ;  often  the  eggs 
in  same  nest  vary  in  depth  of  colouring,  but  there  is  no  rule  as 
to  whether  the  first  or  second  egg  has  most  colouring." 

Two  eggs  seem  to  be  the  usual  number ;  the  pair  are  usually 
alike,  but  those  from  different  nests  vary  very  much — some  are 
almost  round,  others  much  elongated ;  some  blackish  brown,  and 
others  almost  white. 

I  have  known  a  third  egg  laid  in  a  nest  from  which  one  had 
been  abstracted,  one  having  been  left ;  but  whether  the  third 
egg  was  laid  by  the  same  bird  is  of  course  "  not  proven." 

Verner,  on  the  18th  of  April,  1879,  found  Neophrons  laying  in 
an  old  nest  of  the  Snake-Eagle  (Circaetus  gallicus),  from  which  I 
had,  in  May  1877,  shot  an  old  bird ;  this  nest  was  on  a  bough  of 
a  cork-tree,  about  twenty  feet  from  the  ground.  He  took  an  egg 
on  the  18th,  and  on  the  25th  a  second  egg,  snaring  one  of  the 
old  birds,  ultimately  to  liberate  it.  The  Snake-Eagles  constructed 
a  fresh  nest  close  by.  This  is  the  only  instance  which  I  know 
of  the  Neophron  nesting  in  a  tree  in  Andalucia,  but  the  Indian 
Neophrons  usually  do  so. 

The  nest  is  often  easily  accessible  from  below,  and,  placed  on 
a  ledge  of  some  overhung  rock,  generally  at  the  top  of  a  sierra, 
is  composed  of  a  few  dead  sticks,  always  lined  with  wool,  rags, 
and  rubbish — such  as  a  dog's  head,  boars'  tusks,  dead  kittens, 
foxes'  skulls  and  fur,  rotten  hedgehogs,  dead  toads,  dead  snakes, 
skeletons  of  snakes,  lizards,  mummified  lizards,  lizards'  heads, 
carapaces  of  the  water-tortoise,  rotten  fish,  excrement  both  of 
man  and  beast,  bones,  bits  of  rope  and  paper.  In  one  nest  Verner 


NEOPHRON  PERCNOPTERUS. 


155 


found,  among  a  heap  of  filthy  rags,  a  number  of  meal-worms. 
Probably  the  Neophron  had  picked  up  a  bag  with  some  flour  in  it. 
Naturally  from  the  above-mentioned  contents  their  nests  are  most 
offensively  odoriferous. 


NESTING-PLACE  OF  NEOPHRON. 


Neither  Verner  nor  myself  ever  knew  a  Neophron  to  take  a 
lizard  or  any  animal  alive,  and  we  imagine  that  the  heads  &c. 
found  in  their  nests  are  the  remains  of  those  killed  by  other 
animals.  They  are  probably  among  the  foulest-feeding  birds 
that  live,  and  are  very  omnivorous,  devouring  any  animal 
substance,  even  all  sorts  of  excrement ;  nothing  comes  amiss  to 
them.  Sometimes  they  are  seen  feeding  on  the  sea-shore  on 
dead  fish  thrown  up  by  the  tide. 

Neophrons  in  Andalucia,  like  Griffon  Vultures,  usually  roost 
among  rocks ;  but  in  March  and  April,  1894,  Verner  and  mysel 
noticed  some  twenty  collect  every  afternoon  to  roost  on  some 


156  VULTUKID.E. 

cork-trees   on  the   side    of  a  hill  facing  west.      An  intelligent 
cabrero  told  us  this  was  their  usual  habit. 

I  never  in  the  spring  saw  any  Neophrons  in  the  dark  immature 
plumage,  so  we  may  infer  that  in  a  wild  state  they  get  the  adult 
white  dress,  or  nearly  so,  within  eight  months  from  their  birth  ; 
but  in  captivity  I  have  known  them  to  take  three  years  to  assume 
the  white  plumage. 

Adults.  Whitish,  primaries  black  ;  fore  part  of  head  and  neck  bare  of 
feathers  and  yellow. 

Young.  Dark  brown ;  front  of  head  and  neck  dirty  grey.  Length  25 
inches. 

178.  Gypaetus  barbatus  (Linnaeus) .     The  Bearded  Vulture 
or  Lammergeyer. 

Spanish.  Quebrantahuesos. 

This  Vulture  is  without  doubt  found  on  the  mountains  on  the 
Moorish  side  of  the  Straits,  as  it  is  recorded  from  the  Atlas 
ranges ;  but  we  did  not  notice  any  near  Apes'  Hill. 

On  the  Spanish  side  the  Bearded  Vulture  is  well  known  in 
the  sierras  and  used  to  nest  within  a  short  distance  of  Algeciraz. 

Verner  says  that  "  Twenty  years  ago  these  birds  nested 
regularly  not  far  from  Gibraltar,  but  owing  to  persecution  have  of 
late  years  disappeared  or  retired  to  less-frequented  sierras.  The 
four  nests  I  have  climbed  to  were  all  in  caverns  on  isolated  crags 
on  the  rocky  sides  of  the  hills,  and  they  appear  to  prefer  such 
situations  to  the  great  cliffs  frequented  by  Griffons.  The  young 
do  not  leave  the  eyry  till  June,  as  the  late  Crown  Prince  Rudolf 
of  Austria  had  on  board  his  yacht  a  young  bird  about  three- 
fourths  fledged,  taken  by  him  about  the  18th  of  May  near 
Granada. 

"  The  calreros  or  goatherds  near  Tarifa  hold  these  Vultures, 
right  or  wrong,  responsible  for  any  missing  kids  ;  and  owing  to 
this  idea,  and  to  their  habit  of  nesting  in  crags  which  can  often 
be  approached  without  difficulty  to  within  easy  gun-shot,  added 


QYPAETUS  BARBATUS. 


157 


to  the  great  increase  of  guns   carried  lately,  their  disappearance 
in  the  districts  near  Gibraltar  is  easily  accounted  for." 

Other  causes  of  their  decrease  are  attributed  to  poison  laid  for 
wolves  and,  more  than  all,  ornithological  collectors.  The 
Griffons  have  not  diminished  about  the  Vega  de  la  Janda,  so 
poison  can  have  done  little  damage,  though  in  the  provinces  of 
Malaga  and  Granada  many  Vultures  have  been  poisoned. 


NEST  OF  BEABDED  VULTURE. 


The  name  of  "  Quebrantahuesos  "  is  applied  from  their  well- 
known  habit  of  taking  bones  up  to  a  great  height  and  dropping 
them  on  the  rocks,  so  as  to  break  the  bones  small  enough  to 
be  able  to  swallow  the  fragments.  These  bone-breaking  places, 
which  they  regularly  use  on  tops  of  the  sierras,  are  well  known  to 
the  cabreros,  and  one  which  I  examined  did  not  appear  to  differ 
from  any  other  flat  rock.  The  wedge  shaped  tail  of  the  Bearded 
Vulture  is  very  apparent  when  flying  overhead ;  their  flight  and 


158  VULTUEID.E. 

habits  are  those  of  a  Neophron,  and  the  stories  of  their  taking 
live  prey  are  doubtless  erroneous  and  applicable  to  Eagles. 

The  following  notes  on  the  Bearded  Vulture  in  Andalucia,  as 
observed  by  Mr.  Stark  some  ten  years  ago,  have  been  kindly  sent 
me  by  Lord  Lilford,  and  are  here  given  almost  verbatim,  omitting 
of  course  the  names  of  localities : — 

"  In  Andalucia  is  decidedly  common  in  the  Sierra  Nevada,  the 
Alpujarras,  and  all  the  region  between  Granada  and  Jaen.  In  a 
day's  ride  five  or  six  may  be  seen  flying  over  the  hill-sides  or 
gliding  along  the  face  of  a  cliff  or  down  some  ravine. 

"  In  certain  districts  of  the  Sierra  Nevada,  where  the  Griffon 
does  not  intrude,  the  Qnebrantahuesos  is  especially  numerous,  and 
the  goatherds  and  shepherds,  who  look  upon  these  birds  as 
perfectly  harmless  as  far  as  live  kids  and  lambs  are  concerned, 
seldom  molest  them.  They  are  therefore  far  from  shy,  and  may 
generally  be  seen  hunting  round  the  outskirts  of  some  village  or 
farm,  on  the  look-out  for  bones,  offal,  ordure,  both  of  man  and 
beast,  or  for  soft  materials  to  line  their  nests. 

"  In  the  Honda  mountains  they  are  fairly  numerous,  becoming 
scarcer  towards  Gibraltar  and  Tarifa. 

"  On  the  4th  of  February,  1876,  I  found  my  first  nest  of  the 
Bearded  Vulture  in  a  low  cliff  in  the  Sierra  Nevada.  We  had 
for  several  days  previously  seen  one  or  two  of  the  birds  flying 
over  the  steep  hill-side  above  the  village,  and,  suspecting  that 
they  had  a  nest,  watched  them  carefully  and  explored  many  cliffs 
and  rocks  without  effect.  However,  on  the  morning  of  the  4th, 
as  I  was  sitting,  with  an  old  cazador  named  Juan,  on  the  edge  of 
a  low  and  broken  cliff,  not  more  than  sixty  feet  high,  a  Bearded 
Vulture  came  gliding  along  below  us  and  suddenly  disappeared. 
Climbing  down,  we  walked  along  the  base  of  the  rocks  and  finally 
stood  below  a  narrow  horizontal  slit  in  the  cliff,  not  more  than 
fifteen  feet  above  us,  and  Juan  now  asserted  that  the  Quebranta- 
huesos  had  entered  this  hole.  We  shouted  and  threw  in  stones 
for  some  time  without  result,  but  at  last  a  larger  stone  than  usual 


GYPAETUS  BARBATUS.  159 

brought  out  the  Vulture.  The  entrance  to  the  cave  being  very 
narrow,  the  bird,  unable  to  spread  her  wings  at  once,  fell  almost 
on  to  oar  heads  as  we  stood  below ;  then,  gathering  way,  she 
glided  down  a  steep  slope  below  us,  and  finally  fell  dead  to  a 
charge  of  small  shot  from  Juan's  rustly  old  single  barrel.  The 
bird  proved  to  be  a  very  large  and  magnificently-coloured  female, 
measuring  3  feet  8^  inches  in  length  and  from  wing  to  wing  8  feet 
7  inches.  Irides  pale  straw-colour,  sclerotic  membrane  blood- 
red  ;  bill  bluish-horn  ;  feet  lead-colour. 

"We  then  proceeded,  with  the  aid  of  a  rope,  to  inspect 
the  nest,  and  found  that  the  hole  from  which  the  bird  issued 
led  into  a  small  cave,  on  the  floor  of  which  was  a  good- 
sized  heap  of  heather-stalks,  bits  of  esparto-rope,  sheep's-skin 
breeches,  and  an  old  sandal  (alpargata)  ;  a  slight  hollow  lined 
with  sheep's  wool  contained  two  richly  orange-coloured  eggs : 
one  of  these  was  on  the  point  of  hatching,  the  other  addled. 
Some  men  were  blasting  the  rocks  for  lead-ore  about  two  hundred 
yards  off,  and  goats  were  being  constantly  driven  along  the  slope 
beneath  the  nest  apparently  without  disturbing  the  sitting  bird. 
The  goatherd  assured  me  that  the  Quebrantahuesos  never  injured 
his  goats  or  kids,  but  was  an  innocent  bird  ('  pajaro  inocente '), 
and  not  like  an  Eagle. 

"On  the  17th  of  January,  1884,  I  returned  to  this  place  and 
saw  a  pair  of  Bearded  Vultures  sailing  along  the  rocks  a  few 
hundred  yards  from  the  nesting-place  of  1876.  On  my  return  to 
the  village  my  servant  told  me  that  he  had  been  watching  the 
Queb rant alme sos  and  had  seen  a  Cuervo  (Raven)  fly  out  of  the 
rock  and  drive  them  off.  In  the  evening  the  old  cazador,  Juan, 
came  in  and  said  that  there  were  many  Qtiebrantahvesos  still 
about,  and  that  four  years  ago  the  Prince  of  Germany,  i.  e.  Crown 
Prince  of  Austria,  had  visited  the  spot,  and  he  (Juan)  had  taken 
the  Prince  to  two  nests,  from  which  he  shot  a  pair  of  old  birds 
and  took  two  young  from  each  nest.  The  nest  of  1876  was 
examined  but  found  empty. 


100 


VULTUIUD.E. 


KEST  OF  UEARDED  VULTURE. 


NEST  OF  BEARDED  VULTURE.  NEAR  VIEW. 


GYPAETUS  BARBATUS.  161 

"The  next  morning,  January  18th,  as  we  were  passing  along  a 
mule-track  over  a  pass,  we  noticed  first  one  and  then  a  second 
Bearded  Vulture  fly  into  a  small  cave  only  thirty  feet  up  a  cliff 
facing  the  road,  which  is  much  frequented  by  muleteers  and 
labourers  going  to  the  olive-plantations.  From  the  path  below  I 
could  distinctly  see  both  birds — one  sitting  on  the  nest  arranging 
the  lining,  the  other  (the  male)  standing  on  the  floor  of  the  cave. 
Later  on  I  saw  one  of  the  birds  carry  a  large  piece  of  wool  into 
the  nest,  holding  it,  not  as  an  Eagle  would  do,  in  its  claws,  but 
in  its  bill.  For  some  time  I  watched  both  Vultures  beating  to 
and  fro  over  the  mountain-slopes.  They  fly  with  their  beaks  and 
eyes  turned  towards  the  ground,  after  the  manner  of  Terns.  The 
male,  when  on  the  wing,  is  conspicuously  smaller  than  his  mate. 

"  On  the  1 9th  I  sent  Francisco,  a  professional  hunter  of  wild 
bees'  nests  and  a  splendid  rock-climber,  into  the  nest ;  he  reported 
that  it  was  lined  with  clean  sheep's-wool,  but  had  as  yet  no  eggs. 
While  Francisco  was  in  the  nest,  the  female  bird  watched  him 
from  an  adjacent  crag,  but  no  sooner  had  he  reached  the  valley 
below  than  she  returned  and  stayed  in  the  nest  for  a  few  minutest 
On  the  20th  we  went  to  look  at  some  Bearded  Vultures'  nests  of 
last  year :  one  was  only  six  feet  below  the  brow  of  a  cliff,  and 
was,  as  usual,  built  in  a  small  cave,  being  a  small  flattened  mass 
of  dead  sticks  lined  with  pieces  of  esparto-rope,  and  contained, 
as  in  the  nest  of  1876,  an  alpargata  in  the  lining.  Not  far  off 
was  a  nest  of  the  Golden  Eagle  (Aquila  chrysaetus),  a  vast  pile  of 
sticks  on  an  open  ledge  of  the  cliff. 

"On  the  21st  we  went  to  look  at  a  nest  about  a  league  off, 
which  turned  out  to  be  occupied  by  Golden  Eagles ;  but  Juan 
assured  me  that,  although  a  pair  of  Eagles  had  nested  in  it  in 
1880,  the  year  of  the  Prince's  visit,  Quebrantahuesos  had  bred  in 
it  the  year  before.  Francisco  went  into  the  nest  and  found  it 
empty ;  he  had  just  come  up,  and  we  were  talking  to  a  goatherd, 
whose  flock  of  goats,  sheep,  kids,  and  lambs  were  feeding  about 
the  rocks  below,  when  suddenly  a  large  bird  appeared  round  a 

M 


162  VULTUKIDJ3. 

corner  of  the  cliff,  followed  closely  by  two  others.  '  Quebranta- 
huesos ! '  called  out  Juan ;  but  I  saw  at  once  that  they  were  Golden 
Eagles,  and  so  did  the  goatherd,  who  shouted,  with  an  oath, 

*  Son  aquilas  negras ! '  and  ran  down  to  his  flocks,  making  all  the 
noise  he  could  to  scare  the  Eagles  away,  who  passed  us  within 
easy  shot,  going  down  the  ravine,  and  then  doubled  back  as  if 
intent  on  a  kid  or  a  lamb.     Two  of  these  Eagles  were  fully  adult, 
the  other  had  white  on  the  base  of  the  tail,  probably  the  young  of 
the  preceding  year.     I  was  struck  with  their  laboured  napping 
flight,  compared  with  the  easy  gliding,  and  apparently  effortless, 
motions  of  the  Bearded  Vulture. 

"  On  the  30th  we  went  to  inspect  the  nest  of  the  18th,  and 
from  the  valley  below  the  rocks  I  could,  with  my  glasses,  see 
the  tail  and  ends  of  wings  of  the  old  bird  as  she  sat  on  the  nest, 
but  could  not  see  her  head,  which  was  directed  to  the  back  of  the 
cave.  We  then  inspected  two  old  nests — one  was  in  a  hole  in 
detached  stacks  of  sandstone  rock,  the  other,  in  a  round  hole  in  a 
low  cleft  very  easy  of  access,  was  one  of  the  nests  harried  by 

*  El  Principe.' 

"  On  the  31st  Francisco  arrived  with  ropes,  and  while  Juan  and 
he  were  laying  down  the  rope  I  seated  myself  on  a  projecting 
rock  level  with  the  cave  and  about  thirty  yards  from  it.  From 
this  I  could  look  into  the  hole  and  see  part  of  the  nest  and  the 
bill  of  the  sitting  bird.  On  a  steep  slope  of  debris  fallen  from  the 
cliff  above,  and  which  reached  to  within  ten  yards  from  the  cave, 
a  large  flock  of  goats,  attended  by  a  boy,  were  pasturing.  The 
noise  made  by  the  goats  or  the  cries  of  the  goatherd  attracted 
the  attention  of  the  Vulture,  who  occasionally  stretched  her  head 
out  of  the  cave  to  see  what  was  going  on  below,  but  she  did  not 
appear  to  be  the  least  alarmed  ;  presently  some  stones,  displaced 
by  the  rope  above,  came  whistling  down,  and  the  bird,  looking  out 
again,  caught  sight  of  me  for  the  first  time.  She  hesitated  for  a 
moment,  then  launched  heavily  out  of  the  nest,  and,  after  one.  or 
two  heavy  flaps,  sailed  steadily  away  round  an  angle  of  the  cliff; 


FALCONIDJS.  163 

as  she  passed  close  by  me  I  could  see  every  feather  and  the  hairs  in 
her  beard.  In  a  few  seconds  she  reappeared,  and  as  long  as  we 
remained  continued  sailing  overhead  watching  our  movements. 
Francisco  now  found  his  way  into  the  nest  from  above,  and  called 
out  that  there  were  two  eggs  ;  these  he  lowered  to  me  in  a  bag, 
and  they  proved  perfectly  fresh  and  of  a  uniform  dull  yellow-ochre 
colour,  the  colouring  looking  as  if  it  had  been  carelessly  laid  on 
with  a  brush.  This  season  1884  was,  on  the  sierra,  very  mild,  the 
snow-line  fully  1000  feet  higher  up  than  in  1876,  a  severe  spring, 
when  the  Bearded  Vultures  were  about  to  hatch  their  eggs  on  the 
4th  of  February,  so  that  the  weather  apparently  does  not  influence 
their  time  of  laying.  These  nests  were  at  an  estimated  altitude 
of  4500  feet. 

"  In  the  lowrer  ranges  of  the  Sierra  de  Honda,  towards  Gibraltar 
and  Tarifa,  the  Bearded  Vulture  is  not  very  common ;  the 
Griffon  being,  on  the  contrary,  abundant  in  that  district.  The 
majority  of  the  Bearded  Vultures  seen  here  have  been  birds  in 
dark  plumage,  not  fully  adult. 

"  I  have  never  seen  the  Bearded  Vulture  attempt  to  catch  any 
living  creature.  Juan,  the  cazador,  told  me  that  they  sometimes 
took  rabbits  and  partridges  ;  but  the  goatherds,  whom  I  questioned 
on  the  subject,  asserted  that  they  only  feed  on  carrion  and  dung, 
and  I  have  watched  them  walking  about  among  the  female  goats, 
apparently  picking  up  the  placenta?  and  droppings  of  goats,  but 
never  molesting  either  the  mother  or  the  kids. 

"  In  food,  nest,  and  nesting-place  the  Lammergeyer  is  simply  a 
big  Neophron." 

Description.  See  Plate  (frontispiece). 

FALCONIDJE. 

179.  Circus  seruginosus  (Linnaeus).    The  Marsh-Harrier. 
Moorish.  Hedia  (Favier).     Spanish.  Aguilucho,  Eapina. 
"The  most  common  of  the  Harriers  in  Morocco,  this  bird  is 


164  FALCON  ID.E. 

both  resident  and  migratory  in  the  vicinity  of  Tangier.  Those 
which  migrate,  pass  to  Europe  in  February  and  March,  returning 
in  September  and  October.  They  commence  to  breed  late  in 
March.  Their  eggs  differ  very  much  in  shape,  being  sometimes 
round,  sometimes  elongated." — Favier. 

In  Andalucia,  as  in  Morocco,  over  all  low  wet  ground,  the 
Marsh-Harrier  is  to  be  seen  in  vast  numbers,  particularly  in 
winter.  Great  quantities  remain  to  breed,  sometimes  as  many  as 
twenty  nests  being  within  three  hundred  yards  of  one  another. 
The  latter,  loosely  constructed  with  dead  sedges,  vary  much  in 
size  and  depth,  and  are  usually  placed  amidst  rushes  in  swamps, 
but  sometimes  on  the  ground  among  brambles  and  low  brush- 
wood, always  near  water,  though  occasionally  far  from  marshes. 
They  begin  to  lay  about  the  end  of  March,  and  at  that  season  fly 
up  to  a  great  height,  playing  about,  and  continually  uttering  their 
wailing  cry.  The  eggs  are  bluish  white,  and  usually  four  or  five 
in  number ;  they  certainly  vary  in  size  and  shape,  and  are  often 
much  stained.  Like  the  eggs  of  all  the  Harriers  that  I  am 
acquainted  with,  and  many  others  of  the  Accipitres,  when  blown 
and  held  up  to  the  light  they  show  a  bluish  tinge.  On  the  same 
day  I  found  a  nest  containing  only  one  egg,  nearly  ready  to 
hatch,  and  another  with  six  eggs  (three  quite  fresh  and  the  other 
three  hard  sat-on).  I  believe  that,  if  the  first  set  of  eggs  be  taken 
they  lay  again  in  a  fresh  nest,  as  we  found  sets  of  fresh  eggs  as 
late  as  the  2nd  of  May. 

Verner  says  :  "  Many  nests  are  built  on  old  Coots'  and  Purple 
Herons'  nests  among  reeds  eight  or  ten  feet  high,  and  in  three 
feet  of  water.  Irregular  in  time  of  laying,  I  found  in  1875,  on 
May  7,  nest  with  two  fresh  eggs  ;  on  May  10,  nest  with  three 
hard-set  eggs  ;  in  1879,  on  May  7,  five  fresh  eggs,  and  another 
nest-  close  by  with  four  young  whose  quill-feathers  were  shooting 
out." 

The  Marsh-Harrier  is  a  perfect  pest  to  the  sportsman,  as, 
slowly  hunting  along  in  front,  they  put  up  every  snipe  and  duck 


CIRCUS  JERUGIXOSCS.  165 

that  lie  in  their  course,  making  them  unsettled  and  wild.  I  have 
repeatedly  seen  them  flush  Little  Bustards ;  but  these  merely 
flew  fifty  yards  to  the  right  or  left  out  of  the  Harriers'  line 
of  flight,  and  settled  down  again. 

Cowardly  and  ignoble,  they  are  the  terror  of  all  the  poultry 
which  are  in  their  districts,  continually  carrying  off  chickens, 
and,  like  other  Harriers,  are  most  terribly  destructive  to  the  eggs 
and  young  of  all  birds. 

On  account  of  these  propensities,  I  never  let  off  a  Marsh- 
Harrier  unless  it  spoiled  sport  to  fire  at  one.  Sometimes,  when 
at  Casas  Viejas  and  the  snipe  were  scarce,  to  pass  away  the  time, 
we  used  to  lie  up  in  the  line  of  the  Harriers'  flight  to  their 
roosting-places ;  for  they  always  take  the  same  course,  and  come 
evening  after  evening  within  five  minutes  of  the  same  time. 
Upon  one  occasion  a  friend  and  myself  killed  eleven,  and  during 
that  visit  accounted  for  over  twenty.  We  also  upon  every  possible 
opportunity  destroyed  the  nest  and  shot  the  old  ones  ;  but  it  was 
the  labour  of  Sisyphus,  for  others  immediately  appeared.  How- 
ever, there  was  a  visible  diminution  of  their  numbers  about  Casas 
Yiejas.  We  never  found  rats  in  their  nests  or  crops,  and  believe 
they  have  not  the  courage  to  kill  them  :  small  snakes,  frogs, 
wounded  birds,  eggs,  and  nestlings  form  the  main  part  of  their 
prey.  I  have  seen  the  Marsh-Harrier  hawking  over  the  sea 
about  two  hundred  yards  from  the  shore,  where  there  was 
shallow  water,  but  could  not  discover  what  they  were  taking. 

I  do  not  know  whether  it  is  always  the  case  with  the  Harriers, 
but,  as  far  as  my  observation  goes  with  regard  to  the  Marsh- 
Harrier,  it  seems  that  the  males  do  not  sit,  as  I  have  shot,  and 
seen  shot,  many  from  the  nest,  but  never  saw  a  male  killed  flying 
off  the  eggs,  and  have  noticed  that  the  males  only  leave  the 
nesting-places  to  hunt  for  prey.  I  have  also  observed  the  same 
fact  with  Montagu's  Harrier. 

The  very  old  males  have  the  wings  and  tail  ash-grey ;  when 
flying  in  the  sun,  these  parts  appear  almost  white. 


1G6  FALCOXID^E. 

Outer  web  of  fifth  primary  notched  ;  tail  uniform  above. 

Adult  male.  Head  creamy  white,  streaked  with  blackish  brown ;  wing- 
coverts,  secondaries,  tertials,  and  tail  ash-grey. 

female.  Head  huffish  or  creamy  white,  streaked  with  blackish  brown  ; 
creamy-white  margin  on  shoulders  of  wings.  General  colour  dark  brown  ; 
below  chocolate-brown  ;  tail  brown. 

Young  of  the  year.  Dark  brown  ;  head  chocolate-brown.  In  the  next 
plumage  the  head,  chin,  and  throat  vary  from  rufous  to  creamy  white, 
more  or  less  streaked  with  blackish  brown  ;  but  in  first  plumage  some  are 
said  occasionally  to  have  the  head  creamy  white  ;  irides  blackish.  Length 
21-23  inches. 

180.  Circus  cyaneus  (Linnaeus).    The  Hen-Harrier. 

Moorish.  Bou  hasan  (Father  of  beauty).  Spanish.  Cenizo, 
Ave  de  San  Martin. 

According  to  Favier  this  "  is  the  least  common  of  the  Harriers 
near  Tangier,  being  seldom  met  with." 

On  the  Spanish  side  of  the  Straits,  though  a  resident  bird,  the 
Hen-Harrier  is  most  frequently  seen  in  winter ;  but  their 
numbers  fluctuate  greatly.  We  observed  more  in  the  winter  of 
1871-72  than  at  any  other  time,  particularly  about  Casas  Viejas, 
seldom,  however,  coming  across  an  old  male. 

Outer  web  of  fifth  primary  notched. 

Adult  male.  Pale  slate-grey  above ;  throat  and  chest  bluish  ash  ;  upper 
tail-coverts  and  underparts  below  centre  of  breast  white. 

Female.  Above  brown  ;  hind  neck  streaked  with  whitish  ;  below  whitish 
brown  ;  breast  streaked  with  dark  brown  ;  tail  brown,  with  five  bands  of 
darker  brown. 

Young.  Like  female,  but  marked  with  rufous  on  back  ;  and  the  bars  on 
the  tail  are  rufous.  Length  21|  inches. 

181.  Circus  swainsoni,  Smith.    The  Pale-chested  Harrier. 

Favier  states  that  this  species  occurs  on  passage  in  the  environs 
of  Tangier  in  April.  In  the  Norwich  Museum  there  is  a  specimen 
labelled  "  Tangier." 

On  the  Spanish  side  it  is  not  uncommon  in  spring,  and 
doubtless  breeds  near  Seville,  where  Lord  Lilford  was  the  first  to 
obtain  it  in  1872. 


CIRCUS  CINEKACEL'S.  167 

Outer  w^b  of  fifth  primary  notched  ;  upper  tail-coverts  white,  banded  or 
spotted  with  ashy  grey. 

Adult  male.  Above  pale  slate-grey  ;  cJiest  wldte. 

Female.  Much  as  female  of  Montagu's  Harrier,  except  the  grey  spots  on 
upper  tail-coverts.  Length  17-20  inches. 

182.  Circus  cineraceus  (Montagu).    Montagu's  Harrier. 

Spanish.  Cenizo. 

"This  Harrier  passes  to  Europe  in  March  and  April;  but 
some  remain  to  breed  near  Tangier,  where  they  are  nearly  as 
common  as  the  Marsh-Harrier,  being  seen  during  passage  on  all 
sides  in  pairs.  They  nest  on  the  ground,  laying  five  eggs,  which 
vary  much  in  shape,  the  colour  being  bluish  white,  marked  with 
spots  of  clear  blue,  which,  after  the  egg  is  blown,  turn  yellowish." 
— Favier. 

Near  Lixus,  in  Morocco,  at  the  end  of  April,  we  found  a 
regular  colony :  there  must  have  been  fifteen  or  twenty  pair  on  a 
marsh  across  the  river.  We  had  no  time  to  go  round  and 
examine  the  ground,  and  could  not  cross  the  river  at  that  place  ; 
but  we  could  see  with  my  telescope  the  hen  birds  sitting  dotted 
about  the  marsh.  The  males  took  a  particular  line  across  our 
side  of  the  river ;  so  I  shot  three  for  identification. 

In  the  vicinity  of  Gibraltar  this  migratory  Harrier  is  not  often 
met  with,  except  on  passage  ;  I  did  not  observe  them  in  winter. 
Near  Seville  they  are  very  common,  and  dark  specimens,  some  of 
them  complete  melanisms,  are  frequently  procured. 

Like  other  Harriers,  they  are  terrible  egg-destroyers,  but  other- 
wise harmless.  They  can  be  easily  recognized  by  their  smaller 
size,  lighter  and  more  Owl-like  flight,  their  wings  being  longer  in 
proportion  than  other  European  Harriers. 

Outer  web  of//£/t  primary  entire.  Notch  on  inner  web  of  the  first  and 
outer  web  of  the  second  primary  an  inch  beyond  the  tip  of  the  primary- 
coverts. 

Adult  male.  Above  bluish  grey  ;  throat  and  chest  ashy  grey.  Axillaries, 
flanks,  belly,  and  thighs  white,  with  rufous  streaks.  One  Hack  land  across 
secondaries  visible  when  wings  are  closed. 


108  FALCOX1DJ5. 

Female.  Above  brown  ;  below  bufly  whifo,  striped  \\ith  russet. 
Young.  General  colour  dark  chocolate-brown,  lightest  below.     Length 
18-11)  indies. 


183.  Melierax  polyzonus  (Kiippoll).   The  Many-banded  Hawk. 

Obtained  by  Mr.  Drake  at  Mogador  ('  Ibis,'  1869,  p.  153).  The 
Zoological  Gardens  also  had  one  living  a  short  time  back,  which 
came  diivcl  from  Morocco,  and  Lord  Lilford  received  one  from 
Mogador. 

Above  slaty  blue.  Core  ami  legs  vermilion.  Upper  tail-coverts  white, 
barred  with  .slaty  grey.  Length  21  inches. 

184.  Buteo  vulgraris,  Leach.    The  Common  Buzzard. 

Moorish.  Kesir  Eknnli  (Famer).     Spanish.  Arpdlla. 

According  to  Favier,  the  Common  Buzzard  is  seen  in  flights 
on  passage  in  March  and  April,  like  the  Black  Kite.  I  have 
seen  them  myself  crossing  the  Straits  on- the  llth,  15th,  and  2 -1th 
"of  March. 

On  the  Spanish  side  they  are  very  abundant  from  November  to 
the  end  of  February.  We  never  detected  any  remaining  to  breed 
near  ( Jihrultar  ;  but  from  a  nest  in  a  pine-tree,  containing  two 
eggs,  I  shot  one  on  the  29th  of  April  near  Seville. 

In  the  Cork- wood  of  Almoraima  there  are  certain  high  trees 
which  are  the  favourite  resting-places  of  Buzzards.  These  trees  are 
always  chosen  to  command  a  good  look-out,  and  are  used  winter 
after  winter  in  succession ;  if  one  bird  is  shot,  another  takes  its 
place.  They  are  too  lazy  to  annoy  the  sportsman ;  so,  except 
now  and  then  killing  one  for  identification's  sake,  I  never 
molested  them.  I  once  observed  a  Buzzard  feeding  on  the 
carcass  of  a  donkey,  in  company  with  some  Griffon  Vultures. 

This  species  is  one  of  the  most  useful  of  birds,  destroying  vast 
numbers  of  rats,  mice,  voles,  and  moles;  occasionally  they  take 
small  rabbits ;  they  seize  their  prey  on  the  ground,  Avatching 
for  it  from  a  tree  or  rock.  In  one  locality  in  England  they  almost 


BUTKO  DK.SKRTOttUM.  169 

entirely  subsist  on  earthworms.  Yet  this  valuable  \ermin-killing 
bird  has  been  in  the  British  Isles  nearly  extirpated  by  foolish 
and  ignorant  gaine-|treservers. 

Varies  so  much  in  plumage  as  to  defy  description.  Legs  and  toes  short, 
:nid  bare  of  feathers  ;  legs  about,  ij  inches  long. 

Very  old  birds  :irc  sometimes  very  dark  bluish  black  above,  and  only 
slightly  marked  with  light  markings  on  hrea-t. 

Adult*.  Tail  brown,  barred  with  twelve  or  thirteen  bands  of  darker 
brown. 

Young.  Upper  breast  white,  with  only  a  few  spots  ;  throat  brown,  with 
narrow  white  streaks;  tail  ashy  brown,  crossed  \\ith  leu  liars  of  darker 
brown.  Length  20-2;}  inches. 

185.  Buteo  desertorum  (Dan Jin).    Rufous  Buzzard. 

Moorish.  Khabbas  (great  hunter,  mighty  sportsman). 

"  Resident  near  Tangier,  and  found  in  considerable  numbers 
on  all  sides.  Their  food  consists  of  rats,  mice,  snakes,  frogs, 
large  insects,  leverets,  rabbits,  and  chickens.  They  nest  on 
rocks,  laying  two  eggs  (in  March  and  April)  of  a  white  or 
greenish-white  colour,  spotted  with  yellowish  or  reddish  brown  ; 
sometimes  these  spots  completely  cover  the  thick  end  of  the  egg. 
The  males  sit  in  their  turn.  The  irides  are  yellow  ;  the  third 
and  fourth  quill-feathers,  equal  in  length,  are  the  longest  in  the 
wing.  Twenty-four  eggs  of  this  Buzzard  have  passed  through 
my  hands." — Favier. 

This  red-coloured  Buzzard  is,  as  above  stated,  common  in 
Morocco.  On  the  26th  of  April,  1871,  we  found  a  nest  on  the 
top  of  a  very  tall  olive-tree  in  a  santo  or  burial-ground  in  Garbia, 
shooting  both  the  old  birds,  one  off  the  nest,  which  was  like  a 
Kile's  and  lined  with  fresh  olive-twigs  and  rags.  It  contained 
two  eggs  on  the  point  of  hatching;  they  were  of  a  white  colour, 
thinly  marked  all  over  with  very  small,  short,  reddish-black  lines, 
and  were  more  rounded  than  average  eggs  of  either  of  the  Kites, 
though  I  have  seen  eggs  of  both  lied  and  Black  Kite  very  like 
them. 


170  FALCONID.E. 

In  this  santo,  perhaps  two  acres  in  extent,  were  some  of  the 
tallest  wild  olive-trees  I  have  ever  seen,  on  which  were,  besides  the 
Buzzard's  nest,  one  of  the  Common  Kite,  with  young,  two  of 
the  Black  Kite ;  and  in  a  bramble-brake  at  the  edge  was  a  nest 
of  Marsh-Harriers,  with  young.  The  day  before,  we  took  Black- 
Kite's  eggs  quite  fresh,  which  shows  the  relative  time  of  nesting 
of  the  above-named  species. 

We  always  saw  this  Buzzard  in  wooded  districts,  like  our 
Common  Buzzard,  generally  sitting  on  the  bough  of  some  dead 
tree ;  and  this  makes  me  wonder  that  Favier  did  not  mention 
them  as  nesting  on  trees  as  well  as  rocks. 

On  the  Spanish  side  of  the  Straits  I  never  met  with  any ;  nor 
have  I  seen  a  specimen  which  could  be  referred  to  this  species, 
but  Arevalo  mentions  an  adult  female  which  was  killed  near 
Malaga  on  the  5th  of  February,  1873.  Slightly  smaller  in  size, 
the  adults  might  be  recognized  within  a  hundred  yards  or  so  by 
their  reddish  colour ;  but  the  immature  birds,  dead  or  alive, 
could  not  be  distinguished,  except  possibly  by  size,  from  those  of 
the  Common  Buzzard. 

Adults.  Breast,  thighs,  upper  tail-coverts,  and  tail  rufous. 
Young.   Not  to  be  distinguished  from  those  of  B.  vulgaris.      Length 
20-21  inches. 

186.  Aquila  cnrysaetus  (Linnaeus).    The  Golden  Eagle. 

Moorish.  Ogab.     Spanish.  Aguila  ne'gra. 

"  Is  found  on  passage  near  Tangier,  passing  north  in  January 
and  February,  returning  in  July  and  August.  Some  remain  to 
nest  on  rocks  in  March  and  April." — Favier. 

We  found  in  April  a  nest  of  an  Eagle,  apparently  of  this 
species,  on  a  very  high  cliff  near  Jebel  Musa,  opposite  Gibraltar ; 
but  being  unable  to  obtain  the  bird,  we  left  the  nest  untouched. 

On  the  Andalucian  side,  the  Golden  Eagle  is  found  in  the 
sierras,  but  is  not  common  near  Gibraltar,  and  fast  decreasing ; 
one  pair  used  to  nest  at  San  Bartolome.  They  no  doubt  take 


AQUILA  ADALBERTI.  171 

kids  and  lumbs,  and  probably  tbis  crime  is  erroneously  attributed 
to  the  Lammergeyer.  Towards  Granada  and  Alora  these  Eagles 
are  much  more  common,  and  we  saw  them  close  to  Lanjaron,  in 
April  1877. 

Legs  feathered  in  front  to  the  toes,  the  last  joints  of  which  only  are  covered 
by  three  large  scales  (Aguila) .  General  colour  dark  brown  ;  nape  and  hind 
neck  light  brown  ;  thighs  uniform  dark  brown  ;  tail  with  one  greyish  bar 
on  under  surface.  Length  32-36  inches. 

Young.  Basal  half  of  tail  white  ;  base  of  body-feathers  white.  See 
Plate. 


187.  Aquila  adalberti,  R.  Brehm.    White-shouldered  Eagle. 

Spanish.  Aguila  real. 

The  White-shouldered  Eagle  is  stated  by  Favier  to  be  rare 
near  Tangier.  He  calls  it  Aquila  imperialis,  and  gives  a  local 
name  ("  Larnaj  ")  describing  an  adult  bird.  I  have  examined 
Moorish  specimens  in  immature  plumage,  and  seen  what  I 
considered  to  be  this  Eagle  on  the  wing. 

In  wooded  districts  in  the  west  of  Andalucia  this  Eagle  is 
universally  distributed,  being  most  abundant  in  the  Cotos  towards 
Seville  and  about  Cordova,  not  unfrequently  occurring  near 
Gibraltar.  A  tree-nesting  Eagle ;  the  old  bird  sits  very  close, 
but  not  more  so  than  some  other  Raptores.  Three  eggs  is  the 
usual  complement ;  and  these  are  generally  laid  during  the  first 
fortnight  in  March,  being  usually  white ;  they  are  sometimes 
much  spotted  with  reddish  brown,  and  vary  much  in  shape  and 
size.  One  nest  was  lined  with  horse-dung  as  well  as  green  twigs. 
In  the  winter  they  mostly  roost  close  to  their  nesting-places. 
Since  I  first  met  with  this  Eagle  in  1869,  I  find  now,  in  1894, 
that  their  numbers  are  much  reduced,  chiefly  by  or  owing  to 
collectors. 

In  a  nest  built  in  a  pine-tree  near  Vejer,  Mr.  Stark  informed 
me  that  he  found  seven  rabbits,  three  partridges,  and  a  stilt,  all 
fresh.  The  two  young  birds  in  nest  were  both  dead,  apparently 


FALCONID^B. 


killed  by  a  swarm  of  ants  which  inhabited  a  dead  bough  of  the 
tree  on  which  the  nest  was  placed. 


NEST  OF  WHITE-SHOULDERED  EAGLE  (March  28th,  1894). 

Mr.  Stark,  in  1876,  found  that  they  breed  in  immature 
plumage  ;  but  this  is  not  unknown  with  other  Raptores. 

Verner  writes  that  "  this  Eagle  was  not  uncommon  in  the 
Cork-woods  about  twenty  years  ago,  but  they  have  now,  from 
persecution,  retired  to  the  more  unfrequented  parts  of  the  country, 
and  are  to  be  found  where  there  are  any  cork-trees  big  enough 
for  their  nests,  which  usually  are  lined  with  fresh  green  boughs 
of  the  ilex;  but  nests  found  in  1875  and  1877  were  lined  with 
goat's-hair  and  lamb's-wool,  which  I  take  to  be  exceptional  cases. 
The  White-shouldered  Eagle  nests  with  great  regularity,  laying 
three  eggs  by  about  the  8th  of  March,  and  this  number  seems  to 
be  the  regular  complement. 

"It  is  interesting  to  notice  that  the  habit  of  nearly  all  rock- 
nesting  Eagles  having  two  situations  for  their  nests,  never  far 


i 


«     o  3 


AQUILA  K.EVJA.  173 

apart,  and  which  they  occupy  usually  every  alternate  year,  is 
also  regularly  followed  by  this  tree-nesting  species.  In  every 
instance  in  which  I  have  found  a  nest  there  has  been  a  second  in 
a  tree  within  a  few  hundred  yards  of  the  first  one.  The  Snake- 
Eagle  has  the  same  habit,  so  much  so  that  on  finding  an 
unoccupied  nest,  one  has  only  to  search  the  vicinity  for  a  second 
one,  and  usually  with  success." 

When  sitting  on  trees  or  rocks,  the  white  shoulders  of  this 
Eagle  are  very  conspicuous. 

The  old  birds  are  easily  recognized  on  the  wing  from  their 
very  dark  appearance.  The  immature  are  less  easy  to  distinguish  ; 
for  a  long  time  some  of  them  were  thought  to  be  specimens  of 
A.  rapax;  and  I  remember  being  considered  a  heretic  in 
ornithological  matters  for  saying  they  were  young  White- 
shouldered  Eagles. 

The  adults  are  very  dark  brown  on  the  body  and  wings,  except  for  the 
white  patches  on  the  latter,  whence  its  name.  The  immature  birds  are  at 
first  of  a  uniform  reddish  brown,  which  becomes  gradually  lighter.  They 
take,  in  captivity,  three  years  to  show  any  white  in  the  wing. 

Adult.  See  Plate. 

Immature.  General  colour  light  brown  ;  breast  uniform.  Length  31-33 
inches. 

188.  Aquila  nsevia  (J.  F.  Gmelin).    The  Spotted  Eagle  *. 

The  Spotted  Eagle  does  not  appear  to  have  been  obtained  by 
Favier  in  Morocco.  The  only  two  specimens  which  I  have  seen 
from  Andalucia  were  both  from  near  Seville — one  in  spotted 
plumage  (the  same  mentioned  by  Lord  Lilford  and  Mr.  Saunders 
in  '  The  Ibis '),  and  an  adult  male,  killed  on  the  12th  of 
November,  1870,  and  now  in  Lord  Lilford's  collection. 

General  colour  dark  reddish  brown.     Tail  nearly  uniform  brown  above. 
Young.  Wing-coverts  and  scapulars  with  large  oval  tips  of  white  or 
brownish  wThite.     Length  26  inches. 

*  This  term  "  spotted "  is  misleading,  as  only  the  immature  birds  are  so 
marked. 


174  FALCONID^B. 

189.  Aquila  rapax,  Temminck.    Tawny  Eagle. 

This  Eagle  is  included  in  Favier's  list.  I  never  saw  but 
one  Spanish  specimen,  which,  obtained  near  Malaga  in  winter 
of  1877,  is  in  Lord  Lilford's  collection,  and  was  of  the  light 
stone-coloured  plumage  (young  1)  figured  as  Falco  belisarius, 
Levaillant,  jun.,  Exp.  Sci.  Alger.,  Oiseaux,  pi.  2  (1850). 

Little  seems  to  be  known  of  the  plumage  of  this  African  Eagle, 
which  is  much  smaller  than  A.  adalberti,  the  immature  birds  of 
which  species  were  long  supposed  to  be  Tawny  Eagles. 

The  adult  tawny  birds  are  known  to  pair  with  the  light- 
coloured  ones. 

General  colour  from  rich  reddish  brown  to  light  stone-colour.  Length 
26-30  inches. 

190.  Haliaetus   albicilla    (Linnseus).     The   White-tailed    or 

Sea-Eagle. 

Recorded  as  having  been  obtained  near  Cadiz,  and  Lord 
Lilford  says  it  was  apparently  known  to  the  country-people  on 
the  Lower  Guadalquivir. 

Lower  half  of  leg  bare  of  feathers  ;  that  and  the  toes  covered  in  front 
with  large  scales. 

Adult.  Bill  and  iris  light  yellow  ;  head  and  neck  much  lighter  than 
back ;  tail  white. 

Young.  Bill  blackish ;  iris  brown  ;  tail  whitish,  much  mottled  with 
greyish  brown.  Length  28-34  inches. 

191.  Nisaetus  fasoiatus  (Vieillot).    Bonelli's  Eagle. 

Moorish.  Teir  Thum  (Favier).  Spanish.  Aguila  perdicera, 
Aguila  de  las  rocas. 

"  This,  the  most  common  Eagle  near  Tangier,  is  resident  there, 
though  some  migrate  north  in  February,  and  return  in  July. 
They  are  seen  alone  or  in  pairs  hunting  over  a  wide  extent, 
feeding  chiefly  on  hares  and  rabbits;  they  nest  on  rocks  and 
high  trees,  laying  in  March  one  or  two  eggs,  never  more,  of  a 
rather  round  shape,  rough  and  white  in  colour,  with  sometimes 


DESCENT  TO  NEST  OF  BONELLI'S  EAGLE. 


KISAETUS  FASCIATUS.  175 

green  and  bluish  stains.  On  the  29th  April,  1867,  I  took  a  nest 
containing  one  young  female,  which  was  able  to  fly  on  the  1st 
of  July  and  was  very  savage. 

"  They  are  so  voracious  and  plucky  that  I  have  known  two 
instances  in  which  they  allowed  themselves  to  be  caught  rather 
than  give  up  their  prey :  one  was  taken  by  a  Moor  throwing  his 
burnouse  over  the  Eagle,  which  had  struck  down  a  tame  pigeon ; 
the  other  driving  a  fowl  into  some  brambles,  was  caught  before 
it  would  quit  its  prey." — Favier. 

Bonelli's  Eagle  is  found  generally  distributed  as  a  resident  in 
all  of  the  mountain-ranges  of  Andalucia.  I  know  of  the  sites 
of  many  nests,  but,  not  wishing  to  make  them  public  for  the 
benefit  of  dealers,  refrain  from  mentioning  the  exact  localities, 
merely  observing  that  only  one  couple  appears  ever  to  breed  in 
the  same  range  of  cliffs,  each  pair  holding  its  own  district.  A 
pair  nest  annually  at  Gibraltar,  at  the  "  back  of  the  Rock,"  to  the 
south  of  the  signal-station ;  there  are  never  more  than  a  pair, 
though  there  are  four  situations  where  there  are  nests,  one  of 
which  has  not  been  used  for  several  years.  The  nests  are  built 
of  sticks,  and  placed  on  small  ledges  of  the  steep  rock,  with  one 
exception  well  open  to  observation  from  the  signal-station,  where 
I  used  to  spend  many  an  hour  watching  the  old  birds  and  their 
habits.  For  some  years  they  used  two  of  the  nests  alternately 
year  about ;  and  this  is  said  to  be  a  common  habit  of  both  this 
bird  and  the  Golden  Eagle.  The  sergeant  in  charge  of  the 
signal-station,  and  the  signalmen,  one  of  whom  had  been  there 
eight  years,  all  agreed  that  they  never  knew  two  nests  in  one 
season,  or  saw  more  than  one  pair  of  old  birds.  Lord  Lilford 
asked  me  to  try  and  obtain  the  eggs  for  him ;  so  in  1870  I  made 
arrangements,  by  aid  of  the  "  almighty  dollar,"  with  some  men 
who  had  been  goatherds  at  Catalan  Bay,  to  endeavour  to  secure 
the  prize.  They  laid  ropes  down  from  the  top  to  a  bush-covered 
ledge,  which  was  about  two  hundred  feet  above  the  nest ;  thence 
one  man  lowered  himself;  but  unfortunately  the  nest  was  so 


176 


FALCOMD^E. 


overhung  that,  though  he  could  nearly  touch  the  eggs,  he  could 
not  take  them,  so  was  obliged  to,  reascend  unsuccessful.  The 
next  day  we  arranged  with  improved  gear  to  renew  the  attempt ; 
but  a  very  officious  official  kindly  reported  me  to  the  authorities 
as  disobeying  an  ancient  garrison  order  which  prohibits  animals 
and  birds  on  the  Rock  from  being  destroyed  ;  so  I  had  to  eat 
"  humble  pie  "  and  give  the  affair  up  as  a  bad  business.  The 


LAJA  DE  LA  ZAEGA,  NESTING-PLACE  OF  BONELLl's  EAGLE  AND  GRIFFONS. 

following  notes  as  to  the  time  of  nesting  may  be  interesting. 
Sergeant  Munro,  of  the  Royal  Artillery,  in  charge  of  the  signal- 
station,  assisted  me  with  two  or  three  of  the  dates  during  my 
absence  from  the  Rock. 

In  1869  the  Eagles  nested  on  the  lower  site,  about  300  feet 
from  the  base  of  the  Rock,  which  here  ends  on  the  steep  sand 
slope  south  of  the  village  of  Catalan  Bay. 

In  1870  they  used  the  upper  nest,  and  two  eggs  were  laid ; 


INTERIOR     OF    CAVERN     WITH     NEST    QF     BONELLI'S     EAGLE 


NJSAETUS  FASCIATUS.  177 

the  birds  were  sitting  on  the   20th  of  February ;  only  one  was 
hatched. 

In  1871  the  nest  of  1869  was  repaired,  the  birds  beginning  to 
renew  it  about  Christmas  1870  ;  two  eggs  were  laid  by  the  6th 
of  February,  both  of  which  proved  fertile. 

In  1872  the  upper  nest,  that  of  1870,  was  the  favoured  one : 
the  repairs  began  on  the  20th  of  December,  1871 ;  the  first  of  the 
two  eggs  laid  wras  deposited  on  the  5th  of  February.  On  the 
16th  of  March  both  were  hatched,  making  forty  days  occupied 
in  incubation.  Both  birds  sometimes  sat  at  the  same  time  ;  but 
usually  they  relieved  one  another.  They  continually  turned  the 
eggs  over  with  their  bills  ;  and  sometimes,  when  taken,  the  eggs 
bear  marks  of  this  in  the  shape  of  scratches.  The  upper  part  of 
these  nests  was  always  entirely  rebuilt  with  fresh  green  olive- 
boughs,  lined  with  smaller  twigs  of  the  same.  Some  of  the 
boughs  accidentally  dropped  were  afterwards  picked  up  at  the 
foot  of  the  Rock,  gnawed  through  as  if  by  rats.  It  must  have 
cost  the  Eagles  some  time  and  trouble  to  procure  them,  as  olive 
is  very  hard  and  tough. 

In  1873  I  was  not  at  Gibraltar  ;  but  on  my  return  in  1874,  on 
the  24th  of  February,  it  appeared  that  they  had  built  in  a  fresh 
situation  near  the  other  sites,  and  that  two  unspotted  bluish- 
white  eggs,  rather  smaller  than  the  usual  type,  had  been  taken 
the  day  previously  by  the  aid  of  the  same  men  whom  I  had 
employed  in  1870.  This  nest  was  hid  from  view  of  the  signal- 
station  by  a  projection  of  the  rock,  and  was  easily  obtained,  the 
cliff  there  being  less  than  half  the  height  of  that  where  the  nest 
of  1870  is  placed.  In  company  with  the  officers  who  obtained 
these  eggs,  we  took  another  nest  of  Bonelli's  Eagle  at  some 
distance  from  Gibraltar.  It  was  on  some  rocks  where  the 
previous  spring  they  had  had  the  good  fortune  to  take  two  eggs. 
We  found  the  nest  built  in  a  different  situation,  easily  obtained 
by  the  aid  of  a  rope,  and  very  neatly  built  and  lined  with  twigs 
and  leaves  of  the  cork-tree ;  it  contained  two  splendid  eggs, 

N 


178  FALCONID.E. 

beautifully  marked  with  red  streaks  and  spots,  similar  to  those 
taken  in  1873,  and  doubtless  laid  by  the  same  bird.  I  was  in- 
formed that  the  latter  nest  was  lined  with  leaves  of  the  asphodel, 
and  that  the  spoilers  literally  walked  into  the  nest.  I  saw  the 
situation  myself;  and  it  was  certainly  the  easiest  to  reach  that  I 
know  of,  as  they  usually  build  on  the  face  of  steep  cliffs. 

A  nest  found  in  1874  contained  only  one  egg,  which  was 
addled  ;  but  curiously  enough  the  bird  was  sitting  hard  on  this 
rotten  egg,  and  I  succeeded  in  shooting  the  female.  This  nest 
was  in  a  hole,  and  only  about  50  feet  from  the  base  of  the  steep 
cliff  in  which  it  was  placed,  and  was  lined  with  twigs  and  leaves 
of  butcher's  broom  (Ruscus  hypophyllum).  Not  having  enough 
rope  to  lower  to  the  bottom  of  the  rock,  we  had  much  trouble  in 
getting  the  egg ;  however,  we  sent  for  more  rope  and  lowered  it 
down  from  above,  tying  a  sack  full  of  stones  to  the  end  to  prevent 
it  lodging  in  the  rock ;  but  after  securing  the  object  of  our 
labours  from  below,  we  discovered  that  the  rope,  of  which  there 
was  over  400  feet,  had  become  fixed  in  the  rock  about  halfway 
up,  and  no  power  would  move  it.  The  idiotic  Spaniard  whom 
we  had  left  at  the  top,  when  he  found  that  he  could  not  pull  it 
up,  flung  it  down  without  tying  a  stone  to  the  end ;  so  it  caught 
in  several  places ;  and  by  way  of  finishing  he  came  down  to 
where  we  were  sitting,  and,  after  pulling  violently  at  the  lower 
end,  suddenly  let  go,  when  of  course  the  rope  flew  up  and  lodged 
in  the  rock  out  of  reach ;  so  we  had  to  leave  it  dangling  about 
the  cliff  as  a  memorial  or,  rather,  as  a  Spaniard  remarked,  "  un 
serial  de  los  locos  Ingleses" 

The  usual  number  of  eggs  of  Bonelli's  Eagle  is  two,  and  but 
rarely  one ;  the  colour  is  generally  white,  and  I  have  only  seen  a 
few  marked  with  red  and  buff  spots  and  streaks. 

At  Gibraltar,  Bonelli's  Eagle  may  be  often  seen  suspended,  as 
it  were,  in  the  air,  head  to  wind,  apparently  immovable,  like  an 
artificial  kite,  for  sometimes  nearly  two  minutes.  At  this  time, 
when  watched  through  a  glass,  no  movement  of  the  wings  can  be 


KISAETUS  FASCIATUS.  179 

noticed  beyond  an  almost  imperceptible  quivering ;  but  the  legs 
and  feet  are  continually  shifted  as  if  used  to  balance  the  bird. 
When  not  breeding,  they  hunt  together,  one  high  above  the 
other,  suddenly  stooping  down  on  some  luckless  rabbit  or  else 
gliding  off  to  take  up  a  fresh  aerial  station  whence  to  watch  for 
their  prey,  which  seems  to  be  always  taken  on  the  ground. 
They  feed  chiefly  on  rabbits,  but  have  taken  poultry  away  from 
the  signal-station ;  and  Sergeant  Munro  informs  me  that  one  of 
the  Eagles  once  struck  at  and  seized  his  cat,  but  let  it  go  after 
cutting  its  back  open  and  drawing  blood. 

At  Gibraltar,  in  February,  I  watched  two  Eavens  for  a  long 
time  bullying  one  of  these  Eagles,  which  now  and  then  made  a 
futile  dash  at  his  tormentors,  but  at  last  turned  tail,  leaving  the 
Ravens  masters  of  the  situation.  On  another  occasion,  in  the 
same  month,  I  saw  a  Bonelli's  Eagle  flying  about  not  far  from 
the  Osprey's  nest,  when  down  swooped  an  Osprey,  like  a  stone, 
striking  the  Eagle  on  the  back  and  knocking  out  a  lot  of 
feathers.  Shrieking  out,  they  were  bound  together  for  a  few 
seconds,  and  then  separated,  neither  apparently  the  worse  for  the 
encounter,  and  each  flying  off  towards  their  respective  eyries. 
They  were  so  close  as  to  be  within  easy  shot  when  they  "  collided." 
A  young  bird  about  a  month  old  was  bought  from  a  Moor  at 
Tangier,  and  sent  to  me  on  the  18th  of  April ;  but  it  was  so 
wild  and  savage  that  I  thought  it  would  kill  itself,  and  all  that  I 
have  since  seen  alive  have  been  equally  so. 

The  fully  adult  birds  have  a  white  patch  on  the  back  between 
the  wings ;  and  when  viewed  from  above  this  mark  is  very 
apparent  and  will  at  once  identify  the  species  ;  when  below  them 
the  white  appearance  of  the  underparts  and  their  very  powerful 
gliding  flight  distinguish  them.  To  a  novice  they  mostly  resemble 
the  Osprey  when  on  the  wing ;  but  the  latter  has  a  more  flapping 
flight  and  shows  its  whitish  head. 

The  tarsus,  feathered  to  the  feet,  is  very  long  for  the  size  of 
the  bird,  the  thigh  being  still  longer  in  proportion. 

N2 


180  FALCONID^E. 

Adult.  Above  brown  ;  feathers  of  mantle  showing  a  whitish  patch,  owing 
to  only  the  tips  of  the  feathers  being  coloured  ;  below  white,  streaked  with 
black  ;  irides  golden  yellow. 

Immature.  Fulvous  brown. 

Length  26-29  inches,  tarsus  4|.  From  tip  to  tip  of  extended  wings  a 
female  measured  exactly  5  feet. 


192.  Nisaetus  pennatus  (Gmelin).    The  Booted  Eagle. 

Moorish.  Ta-ferma  (Favier). 

"  This  Eagle  is  migratory,  crossing  to  Europe  in  March  and 
April,  returning  in  September;  some  remain  to  breed  in  the 
vicinity  of  Tangier  to  go  south  for  the  winter  with  the  return 
migration,  and  they  are  abundant  when  on  passage.  They 
nest  on  high  trees  in  April  and  May,  laying  from  one  to  three 
white  eggs,  often  much  stained  and  with  a  rough  surface." — 
Favier. 

On  the  Spanish  side  this,  the  smallest  of  the  European  Eagles, 
is,  about  Gibraltar,  entirely  migratory.  Many  were  noticed 
crossing  on  the  24th  of  March.  They  frequent  wooded  districts, 
and  the  most  plentiful  of  the  birds  of  prey  in  the  Cork- wood 
during  the  summer,  when  their  wailing  cry  may  be  heard  all  day 
long.  The  nests  there  were,  without  exception,  on  oak-trees, 
sometimes  completely  hidden  in  ivy.  In  the  Cotos  near  Seville 
they  generally  built  in  pine-trees.  The  same  nest  is  used  year 
after  year ;  if  the  old  birds  be  shot,  next  season  another  pair  take 
possession  to  repair  and  reline  it  with  fresh  green  twigs  of  the 
oak.  Two  is  the  usual  number  of  eggs ;  I  have  known  three, 
but  frequently  only  one;  their  general  colour  is  pale  bluish 
white,  sometimes  stained  or  spotted  with  faint  buff  marks.  The 
earliest  taken  was  on  the  12th  of  April ;  but  about  ten  days  later 
is  the  best  time  to  get  them.  This  Eagle  when  put  off  the  nest, 
instead  of  flying  straight  away,  stoops  down  till  it  nearly  touches 
the  ground,  and  then  flies  away  gradually  rising. 

Verner  writes : — "This  Eagle  varies  somewhat  in  its  time  of 


9. 


NISAETUS  PENXATUS.  181 

nesting.     On  May  25,  1877,  I  took  an  egg  from  a  nest,  replacing 
it  with  a  hen's  egg  ;  the  next  clay  a  second  egg  was  laid. 

"  On  June  27th,  1879,  I  found  a  nest  with  a  young  bird  about 
three  days  old,  and  an  addled  egg  ;  on  visiting  this  nest  on  17th 
July  the  eaglet  was  only  just  shooting  its  quill-feathers.  On 
July  1st  I  found  another  nest  with  two  fully-fledged  young,  and 
on  the  14th  another  with  one  fully-grown  young.  Most  eggs 
that  I  have  seen  were  stained  with  blood  ;  indeed,  it  is  unusual 
to  get  a  clean  specimen.  On  the  17th,  when  climbing  to  take 
the  young  Eagle  from  the  nest,  an  old  bird  dashed  off  the  nest 
and  sat  on  a  bough  within  thirty  feet,  screaming  while  I  was 
securing  the  young.  This  is  the  only  instance  in  which  a  bird  of 
prey  has  ventured  close  to  me  when  at  their  nests. 

"  I  brought  up  several  of  the  young  :  at  first  they  were  bold 
and  fearless,  but  when  full-grown  became  sulky  and  savage,  and 
would  not  take  food  from  me.  I  broke  them  by  starving,  and 
then  putting  food  down  near  them,  would  read  a  book  for  an  hour 
perhaps,  or  more,  when  at  last  hunger  got  the  better  of  their 
temper,  and  they  would  suddenly  dash  at  the  meat,  and,  turning 
their  backs  raise  their  wings  and  make  a  screen  round  their  food, 
within  shelter  of  which  they  tore  it  to  pieces.  By  degrees  they 
grew  tamer  and  could  be  carried,  coming  regularly  to  the  lure.  I 
flew  two  on  Europa  Flats,  but  was  never  certain  of  them.  I  kept 
these,  as  well  as  other  Eagles  and  Kites,  on  blocks  with  jesses 
and  a  swivel  and  leash." 

Their  principal  food,  judging  from  the  examination  of  nests 
and  the  crops  of  specimens,  appears  to  be  young  rabbits.  These 
Eagles  are  easily  recognized  by  their  small  size  when  on  the  wing 
and  by  the  light  colour  of  the  underparts.  A  local  name  which 
I  have  heard  for  them  is  "  Bacallao,"  from  the  fancied  but  far- 
fetched resemblance  in  colour  and  shape  which  they  are  supposed  to 
have  when  flying  overheard  to  that  staple  article  of  Spanish  diet,  a 
split  dried  salt  codfish ;  but  I  may  as  well  mention  that  I  cannot 
help  thinking  this  name  was  fabricated  for  my  special  benefit. 


182  FALCONID.E. 

The  young  birds  generally  are  of  a  uniform  dark  reddish- 
brown  colour,  but  this  is  not  always  the  case. 

Description.  See  Plate.     Length  23-24  inches. 

193.  Circaetus  gallicus  (Ghnelin).    The  Snake-Eagle. 

Moorish.  Tair  el  hesan.  Spanish.  Culebrera  (the  snake-eater), 
Aguila  parda,  Melion. 

"  Migratory.  Some  remain  to  nest  near  Tangier,  building  on 
very  tall  trees  or  rocks,  laying  in  April  or  May  one  egg,  very 
round  in  shape,  though  slightly  smaller  at  one  end,  of  a  white 
colour,  sometimes  marked  with  rusty  spots.  The  males  sit  in 
their  turn ;  the  young  do  not  fly  till  September.  Those  which 
pass  over  to  Europe  cross  in  March  and  April  to  return  in 
October.  Although  not  uncommon  in  the  vicinity  of  Tangier,  it 
is  more  so  than  Bonelli's  Eagle.  They  will  sometimes  allow 
themselves  to  be  killed  on  the  nest  rather  than  desert  their  young. 
Sixteen  eggs  have  passed  through  my  hands." — Favier. 

"  The  Short-toed  " — best  named  the  Snake-Eagle — is  very  com- 
mon both  in  Morocco  and  Andalucia,  frequenting  wooded  districts 
and  the  valleys  of  the  sierras,  being  by  far  the  most  abundant 
Eagle  near  Gibraltar,  except  the  Booted  Eagle  in  some  localities. 
As  far  as  could  be  observed,  they  are  migratory,  as  I  never  saw 
one  in  the  winter  months — but  Rafael  Mena  got  one  near 
Malaga  on  the  10th  of  January,  1882,  a  bird  apparently  of  the 
second  year,  and  the  only  specimen  but  one  obtained  by  him  there 
up  to  that  date, — the  absence  of  their  chief  food  (snakes  and 
lizards)  at  that  season  being  quite  sufficient  to  account  for  their 
departure,  as  the  temperature  at  that  season,  even  in  sunny 
Andalucia,  is  quite  low  enough  to  cause  these  reptiles  to 
hibernate. 

This  Eagle  breeds  about  the  middle  of  April ;  all  nests  I  have 
seen  were  in  cork,  oak,  or  pine  trees,  and  consisted  of  a  mass  of 
sticks,  generally  lined  with  fresh  leaves  and  twigs  of  the  cork- 


CIRCAETUS  GALLICUS. 


183 


NEST  OF  SN AKE-EAGLE  IN  CORK-TREE. 


ANOTHER  NEST  OF  SNAKE-EAGLE  IN  CORK-TREE. 


184  FALCOXID.E. 

tree.  I  found  one  exception  to  this  among  the  ruins  of  the 
ancient  city  of  Lixus  near  el  Arish,  or  Larache,  in  Morocco,  the 
nest  being  built  in  a  thick  mastick  or  lentiscus  bush,  the  base 
of  the  nest  actually  touching  the  ground  on  the  hill-side.  In  this 
instance  there  was  no  want  of  trees  in  the  neighbourhood  to 
account  for  the  nest  being  placed  in  such  an  unusual  situation. 

On  the  24th  of  April  I  shot  the  hen  bird  as  she  flew  out  of  the 
bush.  Had  she  remained  quiet,  probably  the  nest  would  not 
have  been  found ;  it  contained  the  usual  single  large  rough 
white  egg,  slightly  incubated. 

I  never  knew  the  Short-toed  Eagle  to  nest  in  rocks,  as  Favier 
states,  though  have  often  seen  them  perched  on  crags  and  large 
stones ;  but  it  is  now  well  known  that  no  absolute  rule  can  be 
laid  down  as  to  the  breeding  of  many  species  of  the  Diurnal 
Raptores  exclusively  on  rocks  or  trees  ;  they  simply  accommodate 
themselves  to  the  country,  even  nesting  on  the  ground  if  trees, 
rocks,  or  ruins  are  not  available. 

Verner  remarks  that  "  all  the  nests  I  have  seen  of  the  Snake- 
Eagle  were,  with  one  exception  out  of  about  forty,  on  horizontal 
boughs  of  cork-trees,  between  ten  and  twenty  feet  from  the 
ground.  These  Eagles  take  no  pains  to  conceal  their  nest,  but, 
owing  to  the  similarity  of  the  cork-trees  and  enormous  extent  of 
country  covered  by  them,  manage  to  escape  detection  more  often 
than  might  be  expected.  The  nest  is  abnormally  small  for  so 
large  a  bird,  some  not  being  larger  round  than  a  hand-basin,  and 
are  an  almost  flat  platform  of  sticks  with  a  slight  hollow  in  the 
middle. 

"  The  earliest  dates  of  eggs  were :  2Cth  of  March,  1894  ;  ICth 
of  April,  1878,  one  fresh  egg;  21st  of  April,  1878,  and  25th  of 
April,  1879,  incubated  eggs. 

"  On  25th  March,  1877,  I  saw  great  numbers  passing  north  at 
Gibraltar;  and  on  7th  April,  1876,  one  with  a  broken  wing  was 
picked  up  close  to  Europa  Lighthouse." 

This  bird  appears  to  almost   entirely  live  on  reptiles,  and  is 


ASTUR  PALUMBARIUS.  185 

therefore  very  useful  and  to  be  encouraged.  However,  I  have 
known  of  a  dead  Turtle-Dove  in  their  nest,  which  the  Eagle  could 
hardly  have  caught  uninjured. 

The  long  tarsus,  bare  of  feathers,  will  alone  serve  to  distinguish  it  from 
any  other  bird  of  prey  of  its  size  to  be  met  with  near  Gibraltar. 

Adult.  Above  dark  brown.  Breast  pure  white,  streaked  with  black. 
Primaries  banded  below.  Cere,  legs,  and  feet  pale  greyish  yellow  on  a 
bluish  ground-colour ;  iris  yellow  ;  inside  of  mouth  pale  blue.  Length 
26-30  inches. 

194.  Astur  palumbarius  (Linnaeus).    The  Goshawk. 

Moon'sh.  El  boz  (Favier).     Spanish.  Azor,  Gavilan. 

"This  Hawk  is  resident  near  Tangier,  and  is  frequently  seen 
during  passage ;  but  they  are  rarely  met  with  in  winter.  They 
pass  northwards  in  April ;  those  which  breed  nest  in  May.  The 
eggs  are  pure  bluish  white,  often  much  stained  with  yellow.  The 
young  are  so  fierce  that  sometimes  those  in  the  same  nest  will 
kill  and  eat  one  another." — Favier. 

The  Goshawk,  well  known  in  the  wooded  districts  in  Andalucia 
under  the  same  name  as  the  Sparrowhawk,  is  considered  "  muy 
valiente,"  being  said  to  carry  off  partridges  when  they  fall  to  the 
gun :  this  I  know  from  my  own  experience.  I  can  but  consider 
them  rare,  having  only  met  with  the  nest  once,  on  the  15th  of 
May,  1871,  when  I  shot  the  female  bird  as  she  flew  off  the  nest, 
which  was  a  mass  of  sticks  on  an  alder  tree,  about  fifteen  feet 
from  the  ground  or,  rather,  mud,  in  the  thickest  part  of  the  Soto 
Gordo,  in  the  Cork-wood.  The  nest  was  evidently  not  a  new  one, 
and  seemed  to  be  an  old  nest  of  some  eagle  repaired  by  the 
Goshawks.  It  contained  three  eggs  on  the  point  of  hatching, 
stained  yellow  all  over  with  dirt,  so  as  to  resemble  the  eggs  of 
a  Grebe  which  had  been  sat  on  some  time.  On  washing  one  of 
these  eggs,  however,  the  bluish  ground-colour  appeared. 

I  saw  at  Tangier  several  eggs,  stained  in  the  same  manner, 
marked  as  Goshawk's ;  and  until  I  took  their  fac-similes  did  not 
believe  them  to  be  genuine. 


186  FALCONID.E. 

Lord  Lilford  took  a  nest  of  the  Goshawk,  with  three  eggs 
(which  appears  to  be  the  usual  number),  in  the  Goto  Donana,  in 
April  or  early  in  May. 

Above  ashy  brown  ;  thin  white  line  above  ear-coverts  ;  below  white, 
thickly  barred  with  ashy  brown  ;  tail  ashy  brown,  with  four  broad  bars  of 
darker  brown  ;  iris  yellow. 

Young.  Above  brown ;  below  and  under  wing-coverts  light  buff,  streaked 
with  blackish  brown  ;  tail  brown,  with  five  bands  of  darker  brown  ;  iris 
pearly  white. 

Length  :  male  19,  female  23-24  inches. 

195.  Accipiter  nisus  (Linna3us).    The  Sparrowhawk. 

Moorish.  Bou-umeira  takouk  (Cuckoo-Kestrel).  Spanish. 
Gavilan,  Milano  jaspeado  (Marbled  Kite). 

"  Is  resident  in  the  vicinity  of  Tangier,  and  common  during 
passage  in  small  flights,  which  pass  to  Europe  during  February, 
March,  and  April,  returning  in  August  and  September." — 
Favier. 

The  Sparrowhawk  is  resident  in  wooded  districts  near  Gibraltar, 
though  not  in  any  great  abundance,  being  most  frequent  in  winter 
and  during  migration.  Noticed  passing  the  Straits  on  the  28th 
of  March,  and  I  have  dates  of  nests  obtained  on  the  13th  of  May, 
10th  of  May,  and  17th  of  May  in  different  years,  the  first  two 
nests  containing  fresh  eggs ;  all  were  in  tall  trees,  in  the  Cork- 
wood, near  the  Mill.  Verner  took  a  nest  near  Second  Venta, 
with  four  eggs,  on  8th  May,  1878. 

Toes  rather  slender  ;  the  middle  toe  very  much  longer  than  the  others. 

Adult  male.  Above  slate-blue  ;  nape  mottled  with  white  ;  cheeks  and 
ear-coverts  bright  rufous  ;  below  white,  barred  with  bright  rufous  ;  iris 
orange.  Length  12  inches. 

Female.  When  very  old  rarely  assumes  the  same  plumage  ;  the  adult 
females  usually  have  the  breast  barred  with  greyish  brown,  with  a  reddish 
patch  of  downy  feathers  on  the  flanks.  Much  larger  than  males.  Length 
]5  inches. 

Young.  Above  brown ;  nape  mottled  with  white  ;  feathers  of  back  edged 
with  rufous  brown  ;  below  white,  barred  with  rufous  brown  ;  iris  pale 
yellow. 


MILVUS  MIGRANS.  187 

196.  Milvus  ictinus,  Savigny.    The  Red  Kite. 
Moorish.  Siwana.     Spanish.  Milano  real. 

"  Found  in  the  vicinity  of  Tangier  in  much  smaller  numbers 
than  the  next  species,  being  seen  on  passage  only  in  pairs ;  the 
birds  which  remain  to  nest  appear  to  be  those  which  are  the  first 
to  go  south ;  the  remainder  cross  to  Europe  in  March,  returning 
in  October ;  a  few,  however,  stay  throughout  the  winter.  The 
eggs,  two  or  three  in  number,  are  very  similar  to  those  of  the 
Black  Kite,  but  always  larger." — Favier. 

The  Red  Kite  is  resident  and  to  be  seen  almost  everywhere  on 
the  Spanish  side  of  the  Straits — though  in  the  immediate  vicinity 
of  Gibraltar  they  seldom  occur  except  on  passage ;  but  Verner 
found  between  1875-79  two  pairs  nesting  about  the  first  week  in 
April,  in  some  pine-woods  not  very  far  from  the  Rock,  using 
Ravens'  old  nests.  They  are  as  common  in  winter  as  at  any  other 
season,  and  they  particularly  aifect  districts  where  there  are  many 
pine-trees,  on  which,  in  company  with  the  Black  Kite,  they  nest, 
but  from  a  month  to  at  least  a  fortnight  earlier,  and  never  in 
such  numbers  as  that  bird.  The  Red  Kite  is  easily  distinguished 
from  the  next  species,  when  on  the  wing,  by  the  light  colour  and 
much  more  forked  tail ;  when  flying  overhead  by  the  wings, 
which,  underneath,  are  light  coloured,  with  one  dark  patch  on 
each  ;  in  the  Black  Kite  the  underparts  of  the  wing  are  dark. 

General  colour  rufous.  Tail  rufous  and  much  forked  ;  legs  short ;  iris 
yellow.  Length  24  inches. 

197.  Milvus  migrans  (Boddaert).    The  Black  Kite. 
Moorish.  Siwana.     Spanish.  Milano  negro. 

"  Seen  near  Tangier  in  immense  flights,  which  pass  over  to 
Europe  in  February  and  March,  to  return  in  August  and 
September.  Many  remain  to  breed,  awaiting  the  return  migration 
from  Europe,  when  they  all  disappear  for  the  winter." — Favier. 

Though  a  Spanish   name  is  given  above,  very  few  Spaniards 


188  FALCONIDJE. 

distinguish  the  difference  between  the  Common  and  the  Black 
Kite ;  "  black,"  however,  is  a  misnomer,  as  the  primaries  are  the 
only  part  of  the  plumage  which  is  of  that  colour.  The  name 
migrans,  by  which  this  Kite  is  generally  known,  is  most 
appropriate,  as  they  are  entirely  migratory — the  earliest  day  on 
which  I  observed  them  crossing  the  Straits  being  the  5th  of 
March,  then  in  great  numbers,  other  days  on  which  large  flights 
passed  being  the  26th,  27th,  and  28th  of  that  month,  some  on 
the  23rd,  one  on  the  29th  of  April,  and  six  or  seven  on  the  5th 
of  May.  The  latest  date  of  the  return  migration  was  the  9th  of 
October.  Verner's  dates  of  large  flights  passing  are  25th,  31st, 
and  28th  of  March,  in  1877,  '8,  '9.  Meade- Waldo  at  Tangier 
noticed  on  22nd  March,  1892,  quantities  passing  over  to  Spain 
at  the  same  time  as  Cranes  and  Storks.  More  abundant  in  the 
vicinity  of  Seville  and  where  there  are  pine- woods  ;  very  few 
remain  to  breed  near  to  Gibraltar,  but  we  noticed  a  pair  about  the 
western  slopes  of  the  Sierra  Bartolome  on  the  5th  and  6th  April, 
1894.  Both  in  Morocco  and  in  Andalucia  they  nest,  often  in 
colonies,  about  the  end  of  April ;  and  on  the  24th  of  that  month 
we  took  two  nests  near  Larache,  each  containing  the  usual 
number  of  two  eggs,  both  lots  quite  fresh.  The  nests,  built  of 
sticks  and  placed  in  tall  trees  like  those  of  the  Common  Kite,  are 
lined  with  rags,  paper,  bits  of  rope,  and  such-like  rubbish. 

Verner  remarks  that  these  Kites  have  a  very  peculiar  habit  of 
collecting  on  the  open  patches  of  baked  mud  in  the  marisma, 
and  crouching  down  like  Pratincoles  ;  he  saw  as  many  as  twenty- 
two  on  26th  May,  1879,  in  this  position,  and  although  the 
ground  was  like  iron,  and  all  herbage  burnt  up  by  the  sun,  they 
appeared  to  be  eating  something  at  times. 

The  eggs  are  subject  to  great  variation  both  in  shape  and 
colour ;  sometimes  they  are  almost  white,  without  any  spots  ; 
others  are  richly  marked  all  over  with  reddish  brown ;  some  only 
so  marked  at  the  ends,  generally  at  the  large  one,  though  now 
and  then  at  the  smaller  end. 


TERMS  APIVORUS.  ISO 

Above  dark  brown,  below  rufous  brown  ;  bill  black.  Heal  and  throat 
whitish,  with  black  stripss.  Tail  brown,  not  much  forked.  Length  22 
inches. 

198.  Pernis  apivorus  (Linnseus).    The  Honey-Buzzard. 

Moorish.  Khabbas  el  grain  (Favier).  Spanish.  Aguila  de 
Moros. 

"  Only  observed  near  Tangier  during  passage,  migrating  north 
during  April  and  May,  returning  in  August  and  September. 
The  autumnal  migration  is  not  in  such  great  nights  as  the  vernal 
one,  the  greatest  number  seen  in  autumn  being  from  twelve  to 
fourteen,  usually  six  or  eight,  while  in  spring  nights  of  many 
more  than  a  hundred  may  be  seen  crossing  the  Straits  in  a  body. 
Their  plumage  is  so  variable,  it  is  almost  impossible  to  find  two 
exactly  alike." — Favier. 

The  Honey-Buzzard,  as  above  stated,  is  to  be  seen  in  swarms 
during  the  spring  migration,  which  extends  over  some  twenty 
days,  being  at  its  climax  about  the  8th  of  May,  but  many 
hundreds  passed  on  the  12th  of  May  between  Gibraltar  and 
Malaga.  The  latest  flight  noticed  was  on  the  15th  of  that 
month.  When  they  have  once  passed  the  water  the  passage  is 
usually  made  in  a  gyrating  flight  of  eccentric  circles,  sometimes 
very  high  and  as  often  within  shot  of  the  ground.  They  seem, 
when  thus  circling  onwards,  as  if  about  to  alight ;  but  I  never 
saw  them  do  so,  nor  ever  saw  them  except  at  the  period  of 
migration.  Lord  Lilford  observed  large  flocks  passing  south  in 
September. 

We  found  the  Honey-Buzzard  nesting  in  the  province  of 
Liebana  near  Santander,  in  1876 ;  it  would  be  interesting  to 
know  their  most  southern  breeding-range. 

Lores  or  spaces  between  eyes  and  bill  covered  with  feathers  ;  legs  finely 
reticulated  all  round.  (Perms.) 

Adult  male.  Head  ash-grey  ;  above  brown  ;  below  white,  the  chest 
barred  and  spotted  with  brown. 

Adult  female.  Like  male,  but  has  not  the  grey  head. 


190  FALCONID^E. 

Young.  Head  whitish  ;  upper  parts  brown,  marked  with  white  ;  below 
white,  streaked  with  brown. 

A  dark  or  melanistic  form  also  occurs,  in  which  the  underparts  are 
uniform  brown,  the  young  birds  being  striated  with  blackish  brown. 

Length  22-25  inches. 

199.  Elanus  cseruleus  (Desfontaines).   Black-shouldered  Hawk. 
Moorish.  Aislia  hemika  (Favier). 

"  Scarce  in  the  vicinity  of  Tangier,  being  seldom  seen — and 
then  in  very  limited  numbers,  in  February  and  March  and  again 
during  September  and  October.  They  are  more  common  near 
Larache,  where  some  are  found  breeding  in  April.  They  live  on 
birds  and  small  mammals,  and  are  very  voracious.  Their  cry  is 
a  sort  of  whistle." — Favier. 

On  the  African  side  of  the  Straits  we  found  the  Black-winged 
Hawk  common  near  Tetuan  in  April,  as  well  as  about  Cape 
Negro ;  near  Tangier  at  that  time  we  only  saw  two.  They  nest 
on  trees,  and  (as  in  other  countries  in  which  I  have  seen  them) 
keep  to  slightly  wooded  places,  not  frequenting  open  ground. 

Meade-Waldo  says  they  fly  when  it  is  nearly  quite  dark. 

On  the  Spanish  side  this  Hawk  is  very  .rare.  I  never  obtained 
one ;  but  Lord  Lilford  records  a  specimen  (*  Ibis,'  1865,  p.  177) 
as  occurring  near  Seville.  Easily  recognized  on  the  wing  by  its 
greyish-white  colour ;  it  has  a  peculiar  habit  of  hovering  at  about 
thirty  yards  from  the  ground,  with  the  wings  forming  a  sort  of 
V  or  acute  angle  with  the  body,  never  bringing  them  level  with 
one  another  until  it  flies  off"  to  take  up  a  fresh  position.  They 
are  rather  wary  when  thus  engaged  in  hunting  for  their  prey. 

Lores  feathered. 

General  colour  above  ash-grey.  Below  and  axillaries  white  ;  black 
patch  on  shoulders;  iris  crimson.  Length  13  inches. 

200.  Falco  peregrinus,  Tunstall.    The  Peregrine  Falcon. 
Moorish.  Teir  el  hor.     Spanish.  Alcon. 

"  Is  not  uncommon  near  Tangier,  where  some  remain  to  breed ; 


H  CO 


FALCO  PUNICUS.  191 

the  remainder  are  migratory,  going  to  Europe  in'  February  and 
March,  returning  in  November  and  December.  They  nest  from 
March  to  May  on  rocks  and  on  trees,  laying  four  eggs,  eleven  of 
which  have  passed  through  my  hands  for  sale." — Famer. 

The  Peregrine  Falcon  is  common  in  Andalucia  in  winter ; 
those  Falcons  which  are  resident  belong  to  the  subspecies  or 
small  race  called  F.  punicus,  as  no  doubt  do  those  which  Favier 
mentioned  as  nesting  on  the  African  side  of  the  Straits. 

Adult.  Above  bluish  grey,  paler  on  rump,  barred  with  a  darker  tint. 
Crown,  cheeks,  ear-coverts,  and  short  moustachial  band  blackish  ;  below 
white,  with  a  buffy  tinge  on  the  breast,  spotted  on  the  throat  and  upper 
breast,  and  barred  on  the  lower  breast  with  blackish  ;  cere,  eyelids,  and 
legs  yellow. 

Young.  Above  brown,  with  buff  margins  to  feathers  ;  below  whitish, 
with  dark  brown  streaks  ;  tail  irregularly  barred,  tipped  with  whitish  buff; 
cere,  eyelids,  and  legs  blue. 

Length  15-20  inches. 

201.  Falco  punicus,  Levaillant,  jun.  The  Mediterranean  Falcon. 

This  small  race  of  the  Peregrine  is  resident  about  the  Straits  ; 
one  pair  nest  on  the  Rock  near  O'Hara's  Tower,  occasionally 
coming  into  the  town  and  carrying  oif  tame  pigeons.  A  pair 
used  to  breed  close  to  Cape  Spartel,  in  the  middle  of  a  colony  of 
Rock-Doves  ;  but  I  did  not  observe  that  they  molested  their 
neighbours. 

Lord  Lilford  kindly  writes  to  me  as  follows : — "  There  is  no 
specific  difference  between  F.  punicus  and  F.  peregrinus  ;  but  the 
former  is  generally,  if  not  always,  the  smaller  bird,  and  never  in 
my  experience  gets  the  white  breast  and  black  bars  of  the  adults 
of  the  typical  European  race  —  especially  noticed  in  winter- 
killed specimens  from  Egypt,  Italy,  Algeria,  and  Spain.  As  a 
rule,  the  adult  F.  punicus  has  the  breast  more  or  less  tawny,  and 
the  barring  closer  than  in  F.  peregrinus,  but  this  is  not  invariable. 
jP.  punicus,  in  my  opinion,  is  a  good  local  race,  like  F.  anatum 
and  F.  melanogenys  of  America  and  Australia." 


102  FALCONIDJ2. 

In  the  '  Ibis'  for  1887,  p.  275,  plate  viii.,  Lord  Lilford  treats  of 
this  Falcon,  and  considers  that  without  doubt  it  is  "  the  Barbary 
Falcon  "  of  our  old  English  authors  on  Falconry. 

In  the  '  Ibis,'  1882,  pp.  305-321,  the  late  Mr.  J.  H.  Gurney 
gave  an  exhaustive  account  of  the  measurements  and  plumage 
of  F.  barbarus,  F.  punicus,  and  F.  minor.  It  seems  to  me  that 
there  is  no  difference  in  the  immature  plumage  of  the  three 
birds. 

Description.  See  Plate. 

202.  Falco   barbarus,   Linnaeus.     The    Barbary   or  Rufous- 

naped  Falcon. 

The  true  Barbary  Falcon  occurs  near  Tangier,  as  a  specimen 
in  the  Norwich  Museum  came  from  there,  and  I  obtained  another 
from  Olcese. 

This  bird  is,  without  doubt,  identical  with  F.  babylonicus, 
which  shows  how  little  measurements  can  be  relied  on.  Falco 
barbarus  is  figured,  Ibis,  1859,  pi.  vi.,  F.  babylonicits,  Ibis,  1861, 
pi.  vii. 

Adult.  Above  as  in  adult  Peregrine,  but  nape  rufous  ;  lower  parts  with 
rufous  tinge  ;  moustachial  stripe  large.  Length  14-20  inches. 

203.  Falco  lanarius,  Schlegel.    The  Lanner. 

Favier  has,  in  his  MS.,  under  the  head  of  Falco  barbarus^ 
evidently  described  the  Lanner,  as  his  measurements  are  larger 
than  those  of  F.  peregrinns,  instead  of  smaller;  and  all  the 
specimens  of  the  Lanner  which  I  have  seen  from  Tangier,  with 
one  or  two  exceptions,  were  labelled  "  barbarus."  Favier 
adds : — "  This  species,  which  the  Moors  confound  with  the 
Peregrine,  is  resident  and  as  common  as  that  species  around 
Tangier." 

This  bird  seems  to  be  the  most  common  Falcon  in  Morocco. 

On  the  1st  of  May,  1872, 1  obtained  a  female  Lanner  and  three 
eggs.  The  nest  was  on  some  rocks  near  the  above  town.  Two 


FALCO  SUBBUTEO.  193 

of  the  eggs  were  slightly  sat  on  ;  the  third,  much  lighter  in 
colour,  was  addled,  which  is  often  the  case  with  eggs  faintly 
marked  or  differing  from  the  usual  colouring. 

On  the  Spanish  side  of  the  Straits  I  did  not  succeed  in 
obtaining  this  Falcon  near  Gibraltar,  but  have  seen  them  on  the 
wing,  as  has  Verner  in  the  Sierra  Enmedio.  They  were  found 
nesting  on  pine-trees  in  the  Goto  del  Hey,  near  Seville.  In  one 
instance,  an  old  nest  was  used,  from  which  three  years  previously 
I  had  shot  a  Buzzard  (Buteo  vulgaris)  and  taken  two  eggs.  The 
last  nest  found  contained  eggs  at  the  end  of  March. 

Adults.    Forehead   whitish  ;    crown  and   nape  pale  rufous,  with  very 
narrow  black  frontal  line  ;  moustachial  stripe  very  small  and  narrow. 
Young.  Crown  and  nape  lighter.     Length  16-18  inches. 

204.  Falco  eleonoree,  Gene.    The  Eleonora  Falcon. 

I  never  met  with  this  species  on  either  side  of  the  Straits ;  and 
there  is  no  authentic  record  of  a  specimen  having  been  obtained 
in  Andalucia.  Gilbert  White's  brother,  the  Rev.  John  White, 
writing  from  Gibraltar  about  1776,  mentions  the  Hobby  as 
nesting  at  the  "  back  of  the  Rock."  If  a  Hobby  did  nest  there, 
it  could  not  well  have  been  any  species  but  the  Eleonora  Falcon, 
as  the  true  Hobby  (Falco  subbuteo)  is  a  tree-nesting  bird.  The 
Eleonora  Falcon,  however,  occurs  at  Mogador,  and  used  to  breed 
there. 

Adult.  Entirely  blackish  brown.     Wings  as  in  next,  very  long. 

Immature.  Marked  below  somewhat  as  Hobby,  but  lower  parts  more 
rufous  ;  black  moustache.  Tail  much  banded  with  rufous  and  black. 
Length  15^  inches. 

205.  Falco  subbuteo,  Linnaeus.    The  Hobby. 
Spanish.  Alcotan. 

According  to  Favier  this  little  Falcon  is  seen  near  Tangier  in 
pairs  on  passage  only,  "  crossing  to  Europe  in  May,  returning  in 
autumn  to  winter  further  south." 

Near  Gibraltar  the  Hobby  appears  in  the  same  manner ;  the 

o 


194  FALCONID.E. 

earliest  dates  on  which  I  noticed  them  were  the  8th  of  April, 
loth  of  April,  and  20th  of  April,  in  three  different  years,  and  we 
saw  them  near  Seville  very  early  in  May.  They  have  bred  near 
Coria  del  Rio,  and  Arevalo  mentions  them  also  near  Granada. 

The  Hobby  takes  its  prey,  birds  and  insects,  on  the  wing,  and 
no  authentic  record  is  known  of  its  taking  anything  on  the  ground. 

Adult.  Above  bluish  black,  with  black  moustachial  stripe  ;  tbroat  and 
breast  white,  occasionally  tinged  with  buff,  the  breast  striped  with  black  ; 
thighs  and  vent  rusty  red. 

Young.  Buff  edgings  to  feathers  of  buck  ;  thighs  and  vent  only  very 
slightly  rufous.  Length  12-14  inches. 

206.  Falco  eesalon,  Tunstall.    The  Merlin. 

Spanish.  Esmerejon. 

"  Occurs  during  winter  near  Tangier,  coming  from  Europe  in 
September,  returning  north  in  March." — Favier. 

The  Merlin  is  not  uncommon  in  open  ground  in  Andalucia  in 
December  and  January.  The  earliest  noticed  was  on  the  24th 
of  November,  the  latest  on  the  7th  of  March.  About  Casas 
Viejas  they  were  very  plentiful,  and  often  to  be  seen  chasing 
Calandra  Larks ;  they  wrere,  for  the  most  part,  adult  blue-backed 
birds  *. 

Adult  male.  Above  slaty  blue,  with  black  shaft  to  each  feather ;  nape 
rufous  ;  throat  white  ;  underparts  rufous,  streaked  with  blackish  brown  ; 
tail  blue-grey,  the  end  broadly  banded  with  black  and  tipped  with  white. 
Length  10  inches. 

Female.  Extremely  rarely  acquires  the  same  plumage  as  the  adult  male ; 
but  usually  resembles  young  male,  being  above  dark  reddish  brown, 
margined  with  light  brown  ;  nape  whitish,  spotted  with  buff ;  imderparts 
white,  broadly  streaked  with  brown  ;  tail  brown,  crossed  with  narrow 
bands  of  lighter  brown  and  tipped  with  white.  Length  12  inches. 

*  My  friend  Colonel  Delme-Radcliffe,  with  his  vast  knowledge  of  practical 
falconry,  considers  the  Merlin  to  be  one  of  the  most  cowardly  of  the  tribe.  This 
is  rather  against  the  theory  of  book  or  table  naturalists,  that  it  is  one  of  the 
most  courageous  of  the  Falconida},  and  there  is  little  doubt  that  deeds  commonly 
attributed  to  the  Merlin  are  those  of  the  Tiercel  or  male  Peregrine. 


FALCQ  TINNUNCULUS.  195 

207.  Falco  vespertinus,  Liunseus.    The  Western  Red-footed 

Hobby. 

Favier  confounded  this  insectivorous  Falcon  with  the  Hobby, 
calling  it  a  variety;  he,  however,  gives  a  description  which 
identifies  this  species,  and  says  "This  variety  is  found  near 
Tangier  in  April."  They  are  certainly  not  common  there,  and 
said  only  to  appear  when  there  are  locusts,  which  they  follow 
from  the  east.  In  1874,  on  the  27th  of  April,  I  saw  two  near 
Tangier ;  shortly  afterwards  some  were  obtained  by  Olcese ;  and 
just  at  that  time  nights  of  locusts  arrived.  Curiously  enough,  in 
1874,  on  the  very  same  day  in  April  I  saw  one  close  to  Tangier, 
and  the  next  morning  saw  quantities  of  locusts  as  we  were 
crossing  over  to  Gibraltar. 

On  the  Spanish  side  of  the  Straits  one  is  recorded  as  having 
occurred  near  Seville  ;  but  they  are  in  Andalucia  only  accidental 
visitors. 

Claws  yellowish  white. 

Adult  male.  Uniform  lead-grey,  except  reddish-chestnut  thighs,  vent, 
and  under  tail-coverts;  legs  and  feet  red. 

Female.  Above  ash-grey,  barred  with  bluish  black  ;  head  and  nape 
rufous  ;  below  uniform  dull  chestnut. 

Young.  Like  female,  but  head,  nape,  and  underparts  streaked  with 
brown  ;  outside  tail-feather  with  both  webs  barred.  Length  11-12  inches. 

208.  Falco  tinnunculus,  Linnaeus.    The  Common  Kestrel. 
Moorish.  Bou-umcira.     Spanish.  Cernicalo. 

"  Is  both  resident  and  migratory  in  Morocco.  Those  which 
migrate  cross  to  Europe  in  February  and  March,  returning  in 
August  and  September.  They  nest  by  preference  on  old  ruins 
and  walls." — Favier. 

It  is  needless  to  say  much  about  this  Kestrel,  so  well  known 
in  England.  They  are  resident  both  in  Morocco  and  Andalucia, 
and  are  very  common,  more  so  in  autumn  and  spring,  nesting  in 
April  on  trees,  rocks,  and  buildings. 

Adult  male.  Head,  neck,  lower  back,,  rump,  and  tail  blue-grey,  the  latter 

02 


196  FALCOA'JD^E. 

tipped  with  white  on  a  broad  black  band  ;  back  pale  chestnut,  with  small 
black  spots. 

Female  and  young.  Above  entirely  rufous,  banded  with  black;  tail  rufous, 
with  black  bands  and  tipped  with  a  broad  black  band.  Old  females  occa- 
sionally partially  assume  male  plumage,  being  marked  with  blue  on  rump 
and  tail.  Length  13-15  inches. 

209.  Falco  cencriris,  Naumanu.    The  Lesser  Kestrel. 

Moorish.  Souif  (Faoier}.     Spanish.  Primilla. 

"  Is  nearly  as  abundant  near  Tangier  as  the  Common  Kestrel, 
passing  to  Europe  in  February  and  March,  returning  during 
August  and  September." — Favier. 

The  Lesser  Kestrel  is  almost  entirely  migratory,  though  a  few 
remain  at  Gibraltar  during  winter.  Vast  numbers  nest  there, 
chiefly  on  the  steep  face  of  rock  on  the  North  Front.  These 
birds  arrive  about  the  loth  of  February ;  but  I  saw  a  great  flight 
passing  as  late  as  the  4th  of  April.  Probably  these  were  birds 
which  would  breed  much  further  north.  They  nest  on  rocks  and 
ruins,  particularly  on  the  old  Moorish  buildings  and  towers,  of 
which  there  are  so  many  in  Andalucia.  In  some,  as  for  instance 
at  Las  Alcantarillas,  near  Seville,  they  swarm  like  bees  at  a  hive,  as 
also  at  Seville ;  while,  curiously  enough,  at  Cadiz  they  were  absent. 

So  far  as  I  am  aware,  the  Lesser  Kestrel  never  nests  on  trees 
like  the  common  species.  At  the  Goto  del  Key,  on  the  26th  of 
April,  I  took  a  nest  with  four  eggs  out  of  a  hole  in  a  wall  which 
was  within  reach  of  the  ground,  In  the  Crimea,  I  remember, 
they  nested  in  holes  of  river-banks.  On  the  12th  of  May,  near 
Marchena,  we  obtained  sixty  eggs  out  of  an  old  tower,  and  might 
have  taken  as  many  more.  Some  of  these  eggs  were  hard  sat-on ; 
and  the  old  birds  were  caught  on  the  nest,  to  be  released  after 
examination.  The  eggs  varied  very  much,  a  few  being  almost 
colourless,  others  half  white,  half  red,  piebald  in  appearance. 

Adult  male.  Much  resembles  Common  Kestrel,  but  is  smaller,  and  has 
the  back  uniform  pale  chestnut  without  any  spots. 
Female.  Much  as  female  of  last. 
Length  12|  inches.     Claws  white. 


TAXDIOX  HALIAETUS.  197 

210.  Pandion  haliaetus  (Linnaeus).     The  Osprey-. 

Moorish.  Bou  haut  (Father  of  fish).     Spanish.  Aguila  pescador. 

"  This  bird  is  not  uncommon  near  Tangier,  living  among  the 
rocks  on  the  coast,  where  they  nest  in  March,  laying  two  or  three 
eggs ;  the  young  do  not  fly  until  July.  The  migrating  birds 
arrive  in  October  and  November,  returning  north  in  March." — 
Favier. 

The  Osprey  is  most  abundant  in  the  Straits  in  winter.  We 
saw  a  pair  catching  fish  near  Cape  Negro,  at  Lake  Esmir,  in 
April ;  and  a  pair  nest  on  the  rocks  westward  of  Tangier  and  at 
Cape  Negro :  they  also  nest  in  nearly  every  favourable  situation 
on  the  coasts.  Another  pair  regularly  breed  at  Gibraltar,  on  the 
rocks  a  little  to  the  north  of  "  Monkeys'  Cave."  The  Rev.  John 
White  noticed  the  nesting  of  the  Osprey  at  Gibraltar  about  1776  ; 
probably  this  is  the  same  situation,  and  has  been  used  ever  since. 
I  first  knew  of  the  eyry  in  May  1869,  when  there  were  young  in 
the  nest ;  these  did  not  fly  till  the  middle  of  July.  In  1871  the 
nest  was  taken  in  the  middle  of  March,  and  then  contained  three 
eggs  ;  the  old  birds  did  not  leave  the  vicinity,  and  bred  again  the 
next  season,  but  in  a  different  situation  close  to  the  old  one. 
The  first  site  of  the  nest  was  only  to  be  seen  from  the  Europa 
Advance  Battery,  where  I  spent  many  an  hour  watching  the  old 
birds  with  a  telescope.  They,  in  1894,  were  still  at  the  old 
site. 

Being  positive  that  only  one  pair  of  Ospreys  breed  at  Gibraltar, 
and  knowing  the  date  of  laying  of  that  pair,  it  is  difficult  to 
account  for  the  fact  of  seeing,  on  the  23rd  of  April,  one  take  up 
from  the  surface  of  the  sea  and  carry  off  a  stick  or  splinter  some 
three  feet  long ;  and  on  the  30th  of  March  I  also  saw  another 
carrying  a  stick.  Could  this  be  done  in  play  ?  On  the  17th  of 
February  I  saw  one  of  these  Ospreys  give  a  Gannet,  which  had 
ventured  too  near  the  nest,  a  great  buffeting,  knocking  him  about 
and  chasing  him  for  half  a  mile.  The  Isla  de  Palomas,  a  small 
patch  of  rock  near  the  celebrated  and  dangerous  Pearl  Eock,  is  a 


198  PELECANIDuE. 

favourite  resting-place  of  these  birds ;  and  one  is  usually  to  be 
seen  there  at  all  seasons,  perched  on  a  small  pinnacle. 

A  brother  officer  of  mine  killed  an  Osprey  on  the  wing  at 
Europa  Mess-house  with  a  pea-rifle.  The  bird  was  flying  high 
up  over  the  sea ;  but  the  very  strong  westerly  wind  blowing  at 
the  time  caught  and  landed  it  among  the  men's  huts. 

Adult.  Above  brown  ;  below  white,  except  brown  on  breast. 

Young.  Pale  margins  to  feathers  above  :  tail  distinctly  barred. 

Cere  blue.  Legs  and  feet  blue,  very  finely  reticulated  (or  covered  with 
minute  scales)  and  prickly  underneath  ;  legs  very  short,  about  2^  inches 
long ;  outer  toe  reversible  ;  claws  long,  much  curved,  and  very  sharp. 
Length  23  inches. 


Order    S  T  E  G  AN  0  P  OD  E  S. 

Family  PELECANID^l.     Feet  entirely  webbed  or  all 
four  toes  connected  by  webs. 

211.  Phalacrocorax  carbo  (Linnseus).    The  Cormorant. 

Moorish.  Gharrak.     Spanish.  Cuervo  marino. 

"The  Cormorant  is  found  near  Tangier  from  December  to 
February,  and  frequents  the  coasts,  lakes,  and  rivers,  where  it  is 
not  uncommon." — Favier. 

The  above  remarks  equally  apply  to  this  bird  on  the  Andalucian 
side.  I  never  saw  it  in  summer. 

Tail  of  fourteen  feathers. 

Adult.  General  colour  above  purplish  UacTf.  In  spring  slender  white 
plumes  on  head  and  neck  and  white  patch  on  thighs.  These  white  marks 
are  lost  after  breeding-season  is  over.  Length  36  inches. 

Young.   Brown  above,  whitish  below. 

212.  Phalacrocorax  graculus  (Linnaeus).    The  Shag  or  Green 

Cormorant. 

Favier  includes  this  species  in  his  list  as  P.  desmaresti,  stating 
that  it  "is  rare  near  Tangier,  but  found  during  the  whole  year. 


SULA  BASSANA.  199 

The  Shag  is  very  common  and  resident  about  the  Straits.  They 
nest  at  the  island  of  Peregil,  under  Apes'  Hill  on  the  African 
coast,  and  occasionally,  as  in  1883,  at  the  back  of  the  "Rock." 

Tail  of  twelve  feathers. 

Adult.  General  colour  shiny  blackish  green  icithout  any  white.     In  early 
spring  a  crest  curled  forwards  on  front  of  head.     Length  27  inches. 
Young.  Brown  above,  whitish  below. 


213.  Sula  bassana  (Linnaeus) .    The  Gannet  or  Solan  Goose. 

Spanish.  Alcatraz ;  but  this  name  is  often  applied  to  any  large 
Gull. 

Favier  merely  remarks  of  this  bird  that  it  "  arrives  in  October 
and  leaves  during  March,  not  being  very  numerous."  There  are 
always,  however,  during  the  winter  season,  great  numbers  of 
Gannets  in  the  Straits,  particularly  close  to  Gibraltar,  where, 
according  to  the  wind,  they  may  be  noticed  fishing  on  the  lee- 
ward side  of  the  Rock  and  watched  darting  down  from  a  con- 
siderable height  on  their  prey,  often  disappearing  quite  under 
the  wTater.  On  the  wing,  to  an  inexperienced  observer,  they 
appear  like  a  large  Gull.  The  immature  birds  in  their  dull 
spotted  dress,  perhaps  through  not  attracting  so  much  notice, 
seem  to  be  less  in  number  than  the  more  conspicuous  white 
adults  with  their  black  primaries.  The  earliest  dates  on  which 
this  species  was  observed  near  Gibraltar  were  on  the  llth  of 
November,  1870,  and  the  12th  of  October,  1871,  the  latest 
being  on  the  28th  of  March,  1870,  and  the  22nd  of  March,  1871, 
and  the  28th  of  March,  1872.  Many  on  the  28th  of  March, 
1876  (Verner),  one  near  Valencia,  17th  March,  1882,  and  several 
seen  near  Cape  St.  Vincent  on  the  30th  of  April,  1877. 

Adult.  Head  and  neck  huffish  white  ;  the  rest  white,  except  the  black 
primaries  and  primary-coverts  ;  tail  long  and  wedge-shaped. 

Young.  Head,  neck,  and  upper  plumage  blackish  brown,  spotted  with 
white  ;  below  white,  very  thickly  covered  with  blackish-brown  spots,  but 
vary  much.  Length  30  inches. 


200  ARDEID^E. 


Order    HERODIONES.      Family  ARDEID^E. 

214.  Ardea  cinerea,  Linnaeus.    The  Common  Heron. 

Moorish.  Aishush,  Bou-onk  (perhaps  "  ouk,"  neck,  more  likely 
from  cry).  Spanish.  Garza,  Garza  ceniza. 

"This  species  is,  in  the  vicinity  of  Tangier,  both  resident  and 
migratory.  Those  which  migrate  pass  over  to  Europe  during 
February  and  March,  returning  in  November  and  December, 
being  at  all  seasons  plentiful." — Favier. 

The  Common  Heron  visits  the  neighbourhood  of  Gibraltar  in 
great  numbers  during  the  winter  season ;  and  they  particularly 
frequent  the  district  "  between  the  rivers "  near  Palmones. 
Mostly  departing  by  March,  some  few  pairs  are  resident  about 
Casas  Viejas.  I  never  found  them  breeding,  but  some  of  the 
numerous  nests  supposed  to  belong  to  the  Purple  Heron  possibly 
had  the  present  species  for  their  rightful  owner. 

Male.  General  colour  ashy  grey  ;  crown  white  ;  crest  and  nape  black  ; 
hind  neck  pale  grey ;  sides  of  breast  black. 

Female.  Crest  less  developed  and  plumage  duller. 

Young.  Crown  ashy;  sides  of  breast  striped  with  black;  no  plumes  on 
back.  Length  36  inches. 

215.  Ardea  purpurea,  Linnaeus.    The  Purple  Heron. 

Moorish.  Siad  el  mraj  (the  hunter  of  the  marsh).  Spanish. 
Garza. 

"  This  Heron  is,  in  Morocco,  a  summer  visitant,  and  nearly  as 
numerous  as  the  Common  Heron.  They  pass  north  in  April, 
returning  in  September,  many  remaining  in  the  country  to  breed, 
frequenting  reed-beds  and  rushes  on  the  edges  of  lakes  and 
rivers . ' ' — Favier. 

The  Purple  Heron,  in  Andalucia,  only  remains  for  the  nesting- 
season  ;  and  I  never  knew  an  instance  of  its  occurrence  in  winter. 


ARDEA  PURPUREA.  201 

My  earliest  dates  of  arrival  observed  near  Gibraltar  were  the  4th 
of  April  1870,  7th  of  April  1871,  25th  of  March  1872,  7th  of 
March  1874,  25th  of  March  1894.  They  are  extremely  abun- 
dant and  generally  easy  to  get  a  shot  at,  being  seldom  found  in 
the  open,  but  almost  always  among  rushes  or  swampy  jungle, 
and  are  very  rarely  seen  to  perch  on  trees.  There  was,  about 
five  miles  from  Gibraltar,  beyond  the  First  River  (Guadarranque), 
on  the  right  of  the  road  to  Los  Barrios,  a  leech-preserve,  grandly 
called  the  "  laguna,"  perhaps  two  acres  in  extent  and  surrounded 
by  poplar  trees.  This  swamp  was  a  dense  mass  of  tall  rushes 
(Juncus)  springing  up  through  masses  of  dead  ones,  the  growth  of 
years  past,  all  so  matted  and  tangled  together  as  to  make  it 
very  difficult  to  pass  through  them,  more  especially  as  the  water 
was  in  places  up  to  one's  armpits.  This  delightful  spot  was  a 
very  favourite  breeding-place  of  the  Purple  Heron ;  where  there 
generally  used  to  be  three  pairs  nesting,  also  two  nests  of  Marsh- 
Harriers. 

These  Herons  commence  to  lay  about  the  13th  of  April,  as  a 
rule  depositing  three  eggs  (rarely  four),  as  the  following  few 
instances  of  nests  taken  and  seen  will  show: — on  the  21st  of 
April  two  nests — one  with  four,  one  with  three  eggs,  all  fresh ; 
on  the  18th  of  April  two  nests— one  with  one,  the  other  with 
three  fresh  eggs  ;  on  the  6th  May  two  nests — one  with  three  fresh 
eggs,  the  other  with  three  eggs  hard  sat-on.  The  nests,  varying 
much  in  size  and  consisting  merely  of  a  few  dried  rushes 
collected  together  so  as  to  form  a  sort  of  platform  just  clear  of 
the  water,  are  generally  twenty  or  thirty  yards  apart.  The  eggs 
are  light  bluish  green,  similar  to  those  of  Ardea  cinerea. 

It  is  rather  remarkable  that  Purple  Herons  should  generally 
choose  their  building-places  near  to  Marsh-Harriers,  as  the  latter 
repeatedly  rob  them  of  their  eggs.  Many  a  nest  have  I  seen 
with  nothing  but  empty  shells,  the  work  of  the  egg-sucking 
Harrier. 

Adult.  Crown  anl  crest  black,   with  purple   sheen  ;  back,  wings,  and 


202 


AEDEID.E. 


tail  dark  slate-grey  ;  plumes  on  back  pale  chestnut  ;  under  wing-coverts 
chestnut  ;  thighs  rufous  buff  ;  breast  rich  maroon-red. 

Young.  General  colour  above  rusty  red  ;  below  brownish  white.  Length 
30-32  inches. 

216.  Ardea  alba,  Linnaeus.    The  Great  White  Heron. 

We  saw  one  at  the  lakes  of  Ras  el  Doura,  on  the  26th  of 
April,  but  were  unable  to  shoot  it.  I  know  the  bird  well,  having 
shot  them  both  in  the  Crimea  and  in  India.  Mr.  James  J. 
Walker,  R.N.,  in  'Transactions  of  Entomological  Society,'  1890, 
P.  364,  mentions  having  seen  this  Heron  at  Esmir  once  or  twice. 

On  the  Spanish  side  I  never  saw  one,  or  beard  of  a  specimen 
being  obtained ;  but  Verner  saw  one  at  Laguna  de  la  Janda, 
May  18,  1875,  and  Mr.  L.  P.  Irby,  60th  Rifles,  saw  another 
there  in  February,  1894 ;  whilst  they  are  reported  by  Arevalo  to 
occasionally  occur  in  winter  at  the  Albufera,  which  is,  however, 
rather  beyond  the  district  of  "  the  Straits." 

..   Entirely  white  ;  bill  and  iris  yellow  -;  legs  and  feet  black. 

Jn  summer.  Has  bill  black  and  many  very  long  filamented  plumes  on 
back.  Length  36-42  inches. 

217.  Ardea  garzetta,  Linnaeus.    The  Little  Egret. 

Moorish.  Bou-fala,  Bou-bliga,  Bou-biiira  (Favier).  Spanish. 
Garza  blanca. 

"This  bird  is  not  uncommon  near  Tangier  in  small  flights 
when  on  migration.  They  pass  north  in  April,  returning  during 
November  and  December ;  but  some  remain  to  breed  in  the 
country. " — Fa  vier. 

The  Little  Egret  is  the  least  common  of  the  small  Herons 
in  Andalucia,  and,  as  Favier  observes,  some  remain  very  late, 
as  I  have  seen  and  obtained  them  on  the  17th  of  November. 
The  greater  number  arrive  about  the  middle  of  April,  first  date 
observed  13th  April,  1894,  at  Laguna  de  la  Janda,  and  linger 
here  and  there  on  their  route,  gradually  passing  on  to  their 
breeding-places  on  the  borders  of  the  marisma  and  elsewhere. 


ARDEA  BUBULCUS.  208 

They  nest  on  trees,  in  some  seasons,  near  Rocio,  but  are  so 
molested  that  they  change  their  ground  frequently.  When  on 
the  wing,  and  within  a  short  distance,  the  black  legs  and  bill  are 
very  apparent. 

Entirely  white  ;  bill  and  legs  black  ;  iris  yellow  ;  bill  yellow  in  winter. 
In  summer.  Has  long  filamentous  plumes  on  back,  and  two  or  three 
long  feathers  pendent  from  crown.     Length  20  inches. 

218.  Ardea  bubulcus,  Audouin.    Buff-backed  or  Cattle-Heron. 

Moorish.  Tair  el  bukkar  (the  Cow-bird).  Spanish.  Garrapatosa, 
Purgabueyes. 

"This  is  the  most  common  of  the  Herons  around  Tangier, 
and  keeps  in  small  flocks,  always  following  herds  of  cattle,  often 
sitting  on  their  backs,  and  chiefly  feeding  on  insects.  A  small 
proportion  remain-  during  the  breeding-season  ;  but  the  majority 
pass  northwards  in  February,  March,  and  April,  returning  late  in 
the  year," — Favier. 

The  Buff-backed  Heron  is  very  common  in  low-lying  districts 
in  Andalucia,  and  some  are  resident,  breeding  in  the  marisma ; 
but  they  are  very  irregular  in  their  movements,  and  chiefly' 
noticed,  while  passing,  during  March  and  April,  as  they  always 
attend  cattle  when  in  wet  marshy  ground.  The  Spanish  herds- 
men naturally  object  to  have  them  molested,  especially  as  there 
was  at  Casas  Viejas  a  legend  of  a  sporting  Briton  from  Gibraltar 
having  shot  one  as  it  sat  on  a  cow's  back — a  story  which 
unhappily  was  founded  on  fact,  and  only  shows  what  the 
Englishman  is  capable  of. 

The  local  names  of  this  Heron  all  originate  from  its  habit  of 
attending  cattle  and  freeing  them  from  parasites — Garrapatosa,. 
homgarrapata,  a  tick  or  louse:  Purgalneyes,  cattle-cleaner  or 
purifier. 

A  male  bird,  which  had  been  kept  alive  for  about  four  years 
in  the  patio  of  the  Fonda  de  Europa,  at  Seville,  during  the  first 
week  in  April  (his  fifth  spring,  as  far  as  could  be  ascertained) 


204  ARDEID.E. 

began  to  change  the  colour  of  the  legs  and  the  basal  half  of  both 
mandibles  to  a  pinkish  red  ;  the  irides  also  changed  to  beautiful 
rich  pink  colour,  with  a  very  slight  golden  ring  round  the  black 
pupil.  This  change  was  quite  completed  before  the  bird  had 
fully  assumed  the  buff-coloured  back,  which  is  the  mark  of  the 
breeding-dress. 

A  female,  in  confinement  with  the  above-mentioned,  laid  many 
eggs  of  a  very  pale  bluish-white  colour,  showing  a  greenish  tint 
inside  when  held  to  the  light. 

These  captive  Herons  were  quite  masters  of  the  various  Kites 
and  Buzzards  confined  in  the  same  patio,  and  ceaselessly  wandered 
around,  hunting  flies,  which  they  caught  when  settled  on  the 
walls  or  ground,  never  attempting  to  take  them  on  the  wing ; 
but,  poising  the  head  two  or  three  times,  as  a  man  would  a  dart 
before  throwing  it,  they  never  missed  their  aim. 

In  summer.  Crown,  nape,  feathers  hanging  from  lower  neck,  and  plumes 
on  back  reddish  buff  ;  these  feathers  are  elongated  and  hair-like,  especially 
on  the  back  ;  rest  of  plumage  white  ;  bill  reddish  at  base,  yellow  at  tip  ; 
iris  rich  pink,  with  golden  ring  round  pupil. 

In  winter.  Entirely  white,  except  a  patch  of  reddish  buff  on  the  crown  ; 
bill  and  iris  yellow.  Length  18  inches. 

219.  Ardea  ralloides,  Scopoli.    The  Squacco  Heron. 

Moorish.  Aishus  (Favier ;  but  he  applies  this  name  to  all  the 
Herons).  Spanish.  Garza  canaria  (from  colour). 

"  This  species  is  nearly  as  common  around  Tangier  as  the  last, 
occurring  in  small  flocks  during  migration.  Some  remain  in  the 
country  to  breed,  nesting  on  the  ground  among  sedges,  laying  in 
May  and  June  five  eggs,  which  are  more  oval  in  shape  than  those 
of  the  Purple  Heron,  but  of  the  same  colour." — Favier. 

I  found  this  bird  in  great  numbers  about  the  swamps  of  Eas 
el  Doura  towards  the  end  of  April,  where  they  were  by  far  the 
most  common  of  the  Herons. 

On  the  Spanish  side  the  Squacco  Heron  is  entirely  migratory, 
arriving  during  the  month  of  April.  They  are  common  in  the 


NYCTICORAX  GEISEUS.  205 

marisma  of  tlie  Guadalquivir ;  but  I  never  observed  any  near 
Gibraltar,  nor  did  I  ever  see  them  following  cattle,  like  the 
preceding  species.  They  nest  late  in  the  season  ;  but  I  regret  to 
be  unable  to  give  any  personal  information  as  to  their  breeding- 
habits. 

This  species  (beautiful  as  all  the  family  are)  is,  to  my  mind, 
by  far  the  most  handsome  and  elegant  of  all  the  European 
Herons. 

Head  light  buff,  with  crest  of  from  eight  to  ten  elongated,  narrow, 
pointed  white  feathers,  bordered  at  the  sides  with  black  ;  plumes  on  back 
dark  reddish  brown  to  light  buff,  long  and  hair-like  ;  wings,  rump,  tail, 
and  lower  parts,  except  the  buff  feathers  pendent  from  neck,  white  ;  bill 
leaden  blue  at  base,  black  at  tip.  Length  18  inches. 

220.  Nycticorax  griseus  (Linnaeus).    The  Night-Heron. 

Spanish.  Garza  de  noche,  Garza  gris,  Martinete. 

Favier  says : — "  This  species  is  common  near  Tangier  when  on 
migration,  passing  in  small  lots,  which  frequent  wooded  spots 
close  to  lakes  and  rivers."  We  saw  the  Night-Heron  near 
Larache  in  April,  and  near  Tetuan  at  the  end  of  March. 

In  Andalucia  they  are  entirely  migratory,  chiefly  arriving  in 
April ;  but  I  have  no  date  of  their  autumnal  departure,  and  never 
observed  any  very  near  to  Gibraltar.  About  the  district  of 
Seville  they  are  common,  nesting  in  companies  on  trees  on  the 
Rocina  near  Rocio  and  on  the  banks  of  rivers — like  the  other 
smaller  Herons,  breeding  about  the  middle  of  May. 

The  Night-Heron,  as  its  name  implies,  is  a  nocturnal-feeding 
bird,  frequenting  trees  by  day,  and  if  disturbed  usually  flying 
from  one  tree  to  another ;  but  I  have  scarcely  ever  seen  them  on 
the  move  by  day,  unless  frightened  up. 

The  immature  birds,  in  their  brownish  spotted  plumage,  are, 
but  for  their  arboreal  habits,  at  a  little  distance  very  liable  to  be 
mistaken  for  the  Bittern. 

Crown,  nape,  and  back  brownish  black,  with  green  gloss ;  Avings  and 
tail  ash-grey.  Crest  usually  of  three,  but  rarely  of  as  many  as  six  long, 


206  ARDEID.E. 

narrow,  white  feathers  pendent  from  nape  ;  legs  yellow  ;  bill  black  ;  iris 
red.  Females  less  brightly  coloured. 

Young.  Without  crest ;  above  dark  brown,  spotted  with  white  ;  below 
ctiill  white,  striped  with  dark  brown.  Length  21-24  inches. 

221.  Ardetta  minuta  (Linnseus).    The  Little  Bittern.  ' 

"  This  species  is,  near  Tangier,  the  most  scarce  of  the  Ardeid&f 
being  not  often  met  with,  and  then  always  either  alone  or  in 
company  with  the  Squacco  Heron.  They  arrive  and  pass  on 
north  in  April,  and  return  during  August  to  winter  further 
south." — Favier. 

The  Little  Bittern  is,  in  Andalucia,  entirely  migratory,  arriving 
late  in  April.  Considerable  numbers  nest  among  rushes  and 
sedges.  They  are  late-breeders,  nesting  early  in  June,  and  laying 
as  many  as  six  white  eggs.  I  have  no  exact  date  of  the  autumnal 
migration  ;  but  they  are  all  gone  by  October. 

Male.  Crown,  nape,  back,  tail,  wing-quills,  and  upper  surface  of  bill 
black,  with  greenish  gloss  ;  underparts  buff ;  lower  neck-feathers  elongated ; 
feathers  of  upper  breast  blackish  brown,  margined  with  buff ;  iris  yellow. 

Female.  •  Crown  blackish  ;  above  chestnut-brown,  margined  with  buff ; 
primaries  dark  brown. 

Legs  greenish  yellow;  tail  of  ten  feathers.     Length  12  inches. 

222.  Botaurus  stellaris  (Linnseus).    The  Common  Bittern. 

-  Moorish.  Sebaa  el  Mraj  (Lion  of  the  marsh).  Spanish.  Pajaro 
toro  (Bull-bird),  Guia  de  las  Gallinetas  (Guide  of  the  Woodcocks). 

The  Bittern,  according  to  Favier,  "winters  in  Morocco,  and  is 
seen  in  abundance  on  passage,  arriving  during  August  and 
September,  and  leaving  in  February.  They  are  found  in  pairs 
and  in  small  lots,  frequenting  rushes  and  reed-beds."  This  bird, 
.however,  breeds  as  far  south  as  the  neighbourhood  of  Kabat, 
^whence  I  have  seen  the  eggs. 

On  the  north  side  of  the  Straits  they  used  to  breed  at  Casas 
Viejas,  at  the  Laguna  de  la  Janda,  and  in  the  Soto  Torero  near 
.Vejer,  and  still  do  so.  in  the  marshes  pf  Rocio,  near  the  Goto 


207 

Doiiana.  Nesting  about  the  middle  of  May,  they  lay  four  or  five 
pale  brown-coloured  eggs,  placed  in  the  midst  of  thick  rushes. 
I  was  unable  personally  to  find  a  nest,  but  had  several  eggs 
brought  to  me,  and  have  often  heard  them  calling  in  the  daytime 
— a  peculiar,  booming  unmistakable  cry,  whence,  in  almost  all 
countries,  their  local  name  is  derived. 

They  are  more  abundant  in  the  winter  months,  arriving  in  the 
end  of  October,  and  in  some  places  are  at  times  quite  numerous 
wherever  there  are  rushes  and  sedges ;  and  I  have  occasionally 
shot  them  in  the  sotos  of  the  Cork-wood.  Dull  and  sluggish 
in  habit,  it  is  not  until  nearly  trodden  on  that  they  will  rise ;  but 
on  one  occasion  I  remember  finding  several  in  some  rather  open 
marsh,  and  they  flew  up  one  by  one  far  out  of  shot,  seeking 
refuge  in  the  nearest  thick  reed-bed.  Though  often  flushed 
among  sallows  and  bushes  in  the  Soto  Malabrigo,  near  Casas 
Viejas,  I  never  saw  the  Bittern  perch  on  bushes  or  trees  there  OF 
in  any  other  country. 

Crown  and  nape  black;  general  colour  buff;  a,  large  ruff  on  front  of 
neck  ;  the  back  irregularly  marked  with  black  ;  primaries  chestnut,  barred 
ivith  blackish  ;  tail  of  ten  feathers.  Length  28-29  inches. 


CICONIID.E. 

223.  Ciconia  alba,  Bechstein.    The  White  Stork. 

Moorish.  Belarech,  as  pronounced  correctly  "Blarij."  SpatrislL 
Cigiiefia. 

"  This  Stork  is  seen  on  migration  in  vast  numbers  around 
Tangier,  passing  to  Europe  during  January  and  February,  some 
of  the  birds  terminating  their  journey  by  remaining  to  breed  in: 
Morocco.  These  are  the  first  to  depart  south,  returning  again 
year  after  year  to  the  same  places,  and  apparently  by  the  same- 
route  as  that  taken  in  their  gradual  departure., 

"Some  large  flights  pass  on  without  stopping;  those  which 


208  CICON1IDJ1. 

migrate  in  August  rest  awhile  on  their  way  south ;  so  during  the 
autumnal  migration  (which  lasts,  like  the  spring,  for  about  a 
month,  the  latter  half  of  August  and  the  first  part  of  September) 
this  species  is  extremely  numerous  and  seen  around  the  environs 
of  Tangier  in  all  directions ;  they  are  very  tame,  and  often  follow 
close  behind  the  plough. 

"  The  superstition  which  shelters  this  bird  from  molestation 
by  the  natives  has  been  mentioned  in  my  notes  on  the  Swallow  ; 
but  it  may  be  added  that  some  of  the  Arabs  believe  that  the 
Storks  originate  from  a  wicked  Kadi  and  his  family,  who,  as 
a  punishment  for  their  great  cruelty,  were  all  changed  into 
these  birds,  and  that  these  miseralles  humble  themselves  to 
appease  Allah,  and,  in  the  hope  of  some  day  regaining  their 
original  human  form,  pray  without  ceasing  day  and  night,  and 
whenever  they  rest,  prostrate  themselves  and  clack  their  bills." 
— Favier. 

The  White  Stork,  owing  to  the  protection  it  everywhere 
receives,  is  much  more  abundant  in  Morocco  than  in  Audalucia, 
although  plentiful  in  some  level  districts  in  the  latter  country, 
being  most  common  in  the  marismas  and  in  the  vicinity  of 
Seville,  nesting  on  some  of  the  churches  in  that  city.  On  the 
African  side  of  the  Straits,  in  many  situations  they  breed  on 
trees,  generally  in  colonies,  as  well  as  on  houses,  but  usually 
near  villages ;  and  almost  every  Moorish  hovel  has  a  Stork's  nest 
on  the  top,  a  pile  of  sticks  lined  with  grass  and  palmetto-fibre. 
It  usually  contains  four  white  eggs,  which  are  very  rarely  marked 
with  pink  blotches ;  these  are  sometimes  laid  as  early  as  the  25th 
of  March,  and  are  very  good  eating,  either  hot  or  cold.  When 
boiled  hard,  they  have  the  white  clear,  as  with  Peewit's  or 
"  Plover's  "  eggs,  the  yolk  being  of  very  rich  reddish  yellow. 

The  White  Stork  is  rather  irregular  as  to  the  time  of  nesting, 
for  we  found  in  Morocco,  on  the  same  day  (the  25th  of  April), 
young  birds,  eggs,  and  unfinished  nests ;  and,  to  show  how 
varied  is  the  time  of  migration,  saw  on  that  day  a  flight  of  about 


CICON1A  ALBA.  209 

a  hundred,  flying  northward  at  an  immense  height.      As  they 
passed  over  the  "  Storkery,"  which  was  in  a  large  grove  of  high 
trees,  they  lowered  themselves  to  within  a  hundred  yards  or  so  of 
the  nests,  and  after  wheeling  round  a  few  minutes,  as  if  to  see 
how  affairs  were  going  on,  they  worked  up  in  a  gyrating  flight  to 
their  original  elevation,  and  continued  their  northerly  journey, 
doubtless  to  the  great  delight  of  the  resident  Storks,  who  were  in 
a  great  state  of  perturbation  and  disturbance  at  the  appearance 
of  their  brethren.      I   may   here   remark   that    Storks   usually 
migrate  in  large  flocks  at  a  great  height,  with  a  gyrating  flight. 
The  earliest  date  of  their  arrival  that  I  noticed  near  Gibraltar 
•was  on  the  llth  of  January ;  and  they  nearly  all  leave  by  the 
end  of  September.     Feeding  on  insects  of  all  kinds,  mice,  snakes, 
and  other  reptiles,  they  are   most   useful   birds,    and   certainly 
deserve  the  protection  and  encouragement  which  they  receive  in 
Morocco,   where    they   are   in    consequence    excessively   tame. 
Their  grotesque  actions  when  nesting,  and  their  habit  of  con- 
tinually clacking  their  bills  together,  making  a  noise  like  a  rattle, 
render   them   very   amusing  to  watch.      I  was  informed   by  a 
Frenchman  who  had  passed  two  years  in  the  city  of  Morocco, 
that  there,  as  well  as  at  Fez  and  some  other  large  towns  in  the 
Moorish  Empire,  there  is  a  regular  Storks'  hospital,   and  that 
should  one  be  in  any  way  injured,  or  fall  from  the  nest,  it  is  sent 
to  this  institution,  or,  rather,   enclosure,  which  is  kept  up  by 
subscription   from   wealthy   Moors,   wrho   consider   the   Stork  a 
sacred  bird.     I  merely  mention  this  story  to  draw  attention  to 
the   subject  in  case  of  any  future  ornithologists  visiting  these 
cities ;  and  wrere  not  my  informant  worthy  of  credence,  should 
have  omitted  noticing  it. 

White,  except  primaries,  secondaries,  scapulars,  large  wing-coverts,  and 
bare  space  round  eye,  which  are  black.  Bill  and  legs  red.  Length  42 
inches. 


210  TLATALEIDJE. 

224.  Ciconia  nigra  (Linnaeus).    The  Black  Stork. 

Moorish.  Geringa  (Favier  and  D.  Hay).  Spanish.  Cigiiena 
negra. 

"This  species  is  much  less  common  in  the  vicinity  of  Tangier 
than  the  White  Stork  (C.  alba).  They  are  seen  crossing  the 
Straits  during  the  months  of  February,  March,  April,  and  May, 
returning  in  November  to  pass  further  south.  During  their 
passage,  they  keep  in  pairs  and  in  small  lots,  frequenting  much 
the  same  ground  as  the  Crane  (Grus  cinerea)" — Favier. 

From  their  shy  and  wild,  and,  as  far  as  my  observations  go, 
solitary  habits,  the  Black  Stork  on  both  sides  of  the  Straits 
appears  to  be  much  less  common  than  perhaps  is  really  the  case, 
I  saw  one  near  Tangier  in  October  1869,  and  another  on  the 
26th  of  April,  1871,  and  have  seen  several  specimens  obtained  in 
the  neighbourhood.  Verner  reports  seeing  five  daily  near  Tetuan 
between  the  8th  and  12th  of  January,  1890.  Near  Gibraltar  I 
saw  one  on  the  22nd  of  February,  another  on  the  llth  of  January, 
1872,  near  Seville,  and  obtained  a  specimen  from  there  on  the 
18th  of  November,  1870.  These  were  the  only  instances  when  I 
personally  noticed  them. 

Head,  neck,  and  upper  parts  blackish  with  metallic  gloss  ;  only  breast 
and  underparts  white.  Bill,  legs,  and  bare  space  round  eye  red  ;  irides 
reddish  brown. 

Young.  Dark  parts  brownish  black  without  gloss  ;  bill  and  legs  olive- 
green  ;  feathers  of  head  and  neck  tipped  with  buffy  white.  Length  39-40 
inches. 

PLATALEID.&. 

225.  Platalea  leucorodia,  Linmeus.    The  White  Spoonbill. 

Moorish.  Bou-kar-kaba  (Favier)  (Father  of  the  cymbals,  "  kar- 
kaba  "  being  large  wooden  cymbals  used  by  negroes).  Spanish. 
Espatula,  Paleton,  Paleta,  Patera,  Pilato,  Cuchareta. 

"  This  species  occurs  near  Tangier  when  on  passage.     They 


1B1DIDJE.  211 

migrate   north   in   March,   April,    and    May,    returning   during 
October,  and  are  never  observed  in  winter." — Favier. 

We  saw  many  Spoonbills  in  April  at  the  lake  of  Meshree  el 
Haddar,  near  Larache  ;  and  they  then  appeared  to  be  on 
migration. 

The  earliest  occurrence  of  this  species  in  spring  near  Gibraltar 
that  I  know  of  was  one  shot  on  the  9th  of  April,  at  the  First 
River;  and  the  latest  seen  was  a  single  bird  wading  about  the 
river  Barbate,  near  Casas  Viejas,  on  the  20th  of  November.  They 
were  common  in  the  marisma  in  flocks  in  May;  in  some  wet 
seasons  they  nest  there,  and  also  in  the  Soto  Torero,  near  Vejer, 
where,  sad  to  relate,  a  Spaniard,  in  1873,  took  upwards  of  seventy 
eggs  early  in  May.  He  brought  most  of  these  eggs  into  Gibraltar, 
to  some  collectors  who  were  there  at  that  time  ;  and  next  year  he 
described  to  me  the  nests  as  merely  made  of  a  few  sedges,  and 
placed  close  (junto)  together,  each  containing  four  eggs.  The 
season  of  1874  was  very  dry,  and  no  Spoonbills  appeared  there ; 
indeed,  had  it  been  wet,  probably  after  being  so  robbed,  the  birds 
would  not  have  nested  again  in  that  spot ;  but  in  1875  Major 
Verner  reports  them  again  nesting  there  and  also  breeding  during 
the  following  season  in  the  Soto  Malabrigo.  He  further  reports 
seeing  at  La  Janda  a  flock  of  about  a  dozen  on  the  20th  of 
February,  1876. 

White  ;  lower  neck  and  the  short  bushy  crest  tinged  with  yellow  ;  bill 
black  at  base,  yellow  and  broadly  flattened  at  the  end  ;  chin  bare  and 
yellow  ;  iris  dull  red.  Length  32—36  inches. 

Young.  "Without  crest  or  yellow  skin  ;  bill  blackish  ;  iris  black/ 


IBIDID^. 

226.  Ibis  comatus,  Ehrenberg.    The  Red-cheeked  Ibis. 

Favier  included  in  his  MS.,  without  any  description,  another 
species  of  Ibis  as  having  once  been  obtained  by  him  near  Tangier. 
He  called  it  "  His  caha  " ;  but  it  could  hardly  have  been  that 

r2 


212  IBIDUXE. 

South-African  species,  and  was  doubtless  this  rock-frequenting 
and  migratory  bird,  which  has  occurred  at  Mogador,  and  is  said 
to  have  nested  among  the  rocks  on  the  island  there. 

General  colour  iridescent  green,  with  feathers  of  neck  elongated  and 
pointed.     Bill,  head,  bare  throat;  and  legs  red.     Length  24  inches. 


227.  Plegadis  falcinellus  (Linn^us).    Trie  Glossy  Ibis. 

Moorish.  Maiza  (Favier},  Maiza  el  Wad  (D.  Hay)  (river  goat). 
Spanish.  Morito. 

"  This  bird  occurs  near  Tangier  on  passage,  returning  to  pass 
the  winter  further  south.  Some  must  remain  to  nest  in  the 
country ;  for  they  are  frequently  met  with  during  May,  June,  and 
July." — Favier. 

We  saw  great  flocks  of  the  Glossy  Ibis  at  the  lakes  of  Eas 
el  Doura  towards  the  end  of  April ;  but  they  were  very  wary,  as 
they  are  in  Andalucia,  whereas  I  remember  in  India  we  used  to 
walk  up  to  within  thirty  yards  of  them. 

When  flying  they  much  resemble  the  Spoonbill  in  their  manner 
of  flight.  They  nest  in  Morocco,  as  I  have  seen  eggs  obtained 
in  the  country;  and  in  wet  seasons  they  breed  in  the  Soto 
Torero,  near  Vejer,  and  also  in  the  marisma  of  the  Guadalquivir ; 
but  I  have  no  personal  knowledge  of  their  nesting-habits.  The 
eggs  are  of  a  uniform  pale  bluish-green  colour.  Near  Gibraltar 
I  have  only  noticed  this  Ibis  when  passing  late  in  April  and 
in  May.  A  female  shot  at  the  First  River,  on  the  31st  of  May, 
had  the  gizzard  full  of  minute  shrimps ;  four  eggs  in  the  ovary 
were  slightly  enlarged,  which  tends  to  confirm  what  I  have  heard, 
that  they  are  late  breeders. 

Bill  long,  curved  down  (as  in  Curlew)  ;  face  bare  ;  general  colour 
chestnut-brown ;  the  back  much  glossed  with  green  and  purple  ;  legs 
greenish  grey.  Length  22  inches. 


.PHffiNICOPTERIDvE.  213 

Order  0 D 0 N T 0 GL 0 S S -E. 

Family  PHCENIC  OPTERIDJE. 
228.  Phcenicopterus  roseus,  Pallas.    The  Flamingo. 

Moorish.  Nihof.     Spanish.  Flamenco. 

Favier  says : — "  The  Flamingo,  near  Tangier,  passes  north- 
wards in  April,  May,  and  June,  returning  in  August  up  to  as  late 
as  December.  The  females  are  the  first  to  arrive  during  the 
autumn  migration.  The  males  rejoin  their  mates  in  November, 
accompanied  by  the  young  of  the  previous  year ;  the  young  of 
the  year  are  never  seen  here.  They  are  met  with  in  large  flocks 
on  the  lakes,  always  staying  in  the  water,  though  they  never 
swim  about,  and  are  very  wary  and  difficult  to  approach.  The 
only  month  in  which  they  are  entirely  absent  is  July.  Their 
temporary  absence  during  other  months  is  regulated  by  the 
quantity  of  water  in  the  lakes ;  and  as  one  month  is  not  suffi- 
cient time  for  them  to  lay  and  hatch  their  eggs,  they  ought  to 
nest  not  far  from  Tangier:  indeed  an  old  chasseur,  worthy  of 
belief,  informed  me  that  he  had  shot  one  which,  when  it  fell, 
dropped  an  egg  in  the  water." 

The  movements  of  the  Flamingo  are  certainly  very  irregular 
and  perplexing,  and,  no  doubt,  influenced  by  the  amount  of  water 
in  the  brackish  lagoons  which  they  frequent.  Most  of  these 
lagoons,  being  formed  by  rain-water,  are  brackish  from  the  salt 
contained  in  the  earth,  and  in  very  dry  seasons  hold  hardly  any 
water. 

In  very  wet  seasons  these  birds  breed  in  the  marismas  of  the 
Guadalquivir,  and  used  to  nest  at  Fuente  Piedra  between  Cordova 
and  Malaga. 

Flights  of  Flamingoes  are  frequently  seen  passing  near  Gibraltar 
as  early  as  the  4th  of  February,  and  as  late  as  the  1st  of  May ; 
and  they  again  appear  in  August  and  September,  when  immature 
birds  are  met  with.  I  have  seen  flocks  of  thousands  in  the 


214  PH(ENTICOPTERID^E. 

marisma  near  the  Isla  Menor,  and,  by  the  aid  of  a  stalking-horse, 
managed  to  shoot  five  at  a  shot.  Usually  they  are  extremely 
wild  and  shy,  except  during  actual  passage,  when  they  alight  to 
rest  at  the  mouths  of  rivers. 

Their  note  is  not  unlike  that  of  the  Grey-lag  Goose,  and  more 
than  once  at  night  I  have  mistaken  the  sound  for  that  of  these 
Geese. 

Verner  sends  the  following  extract  from  his  journal  for  the 
year  1879  : — "  On  the  27th  of  May,  whilst  in  the  marismas  of  the 
Guadalquivir,  I  made  an  expedition  with  Crown  Prince  Rudolf  to 
obtain  some  Flamingoes'  eggs,  of  which  he  wanted  some  Spanish 
specimens.  After  a  long  and  tedious  ride  across  the  waste  we 
finally  left  hard  ground  and  traversed  the  mud-flats ;  these  were 
either  sun-baked  or  sticky,  and  for  the  last  two  miles  or  so  were 
mostly  covered  with  shallow  water.  The  mud,  where  not  dry, 
was  very  slippery,  and  two  of  our  party  (including  myself)  took 
a  mud-bath,  owing  to  our  horses  slipping  and  falling. 

"  After  what  seemed  an  endless  journey  under  the  broiling 
sun,  we  sighted  an  enormous  flock  of  some  thousands  of 
Flamingoes,  at  where,  the  Spaniards  said,  was  their  usual  nesting- 
station.  When  at  a  good  distance  from  them  they  took  wing, 
except  a  few  which  remained  dotted  about,  and  apparently 
crouching  down  on  the  mud.  The  day  was  extremely  hot,  and 
the  usual  mirage  was  intensified  by  the  steamy  exhalations  from 
the  fast-drying  mud.  Through  this  shimmering  atmosphere  the 
great  flock  of  Flamingoes  on  the  wing  presented  an  extraordinary 
sight,  resembling  a  drifting  cloud,  which  seemed  to  change  colour 
from  a  rosy  pink  to  snowy  white  as  the  mass  of  birds  wheeled  and 
turned  in  their  flight.  On  reaching  the  spot  where  the  birds 
had  remained  behind  when  the  main  flock  flew  off,  we  found  a 
few  eggs  (of  which  I  myself  took  two)  lying  on  small  mud- 
hillocks  raised  above  the  water  and  slush ;  these  eggs  were 
fresh,  white,  tapered  at  both  ends,  and  with  the  chalky  surface 
of  a  Cormorant's  egg. 


AKATID^E.  215 

"  Unluckily,  I  was  prevented  from  visiting  the  main  colony  of 
the  Flamingoes,  for  the  Crown  Prince,  having  got  the  eggs  he 
wanted,  decided  to  return  at  once  to  where  we  had  seen  Lams 
gelastes  in  the  morning,  which  bird,  according  to  Dr.  Brehm,  who 
accompanied  us,  was  not  known  then  to  breed  in  Spain. 

"  Two  wild-looking  men,  frequenters  of  the  marisma,  who  the 
Guardias  contemptuously  termed  Indians,  because  of  their 
savage  appearance,  and  who  had  conducted  us  to  the  Flamingoes' 
nesting-place,  assured  me  that  if  I  would  go  on  for  a  media  legua 
they  would  show  hundreds  of  nests  placed  together  (junto}.  They 
described  the  nests  as  all  small  mud-hillocks,  raised  a  few  inches 
above  the  surface  of  the  water  when  first  built,  but  subsequently 
left  high  and  dry  as  the  water  receded  both  from  evaporation  and 
natural  drainage.  These  men  were,  of  course,  well  acquainted 
with  the  camels,  whose  presence  in  the  marisma  they  seemed  to 
think  the  most  natural  thing  in  the  world,  since  they  had  alike 
been  born  and  bred  in  this  extraordinary  wilderness." 

Mr.  Abel  Chapman  appears  to  have  visited  this  identical 
situation  on  llth  May,  1883,  and  gives  an  interesting  account 
('  Ibis,'  1884,  p.  66,  pi.  iv.)  of  the  Flamingoes'  nests,  but  which 
on  that  date  did  not  contain  eggs. 

Head,  neck,  and  throat  black  ;  front  and  sides  of  head  and  upper  throat 
white  ;  black  mark  between  eye  and  bill.  Length  25  inches. 

Young  in  first  plumage.  Basal  half  of  upper  bill  blackish  ;  general  colour 
greyish  white,  each  feather  above  striated  with  dark  brown  ;  no  pink  in 
the  plumage  ;  legs  and  feet  blackish. 


Order   ANSERES.  Family  AN ATID.33. 

Subfamily  CYGNIN^E. 

229.  Cygnus  musicus,  Bechstein.    The  Whooper  Swan. 
Spanish.  Cisne. 

This  is  the  only  species  of  Swan  which  I  was  able  to  identify 
in  Andalucia,  having  examined  one  specimen  shot  on  the  Guadal- 


216  ANATHXE. 

quivir  below  Seville,  where  in  some  winters  they  occasionally 
occur. 

Adults.  White  ;  bill,  basal  part  to  below  nostrils  yellow,  the  rest  black. 
Young.  Greyish  brown  ;  bill  flesh-colour.     Length  60  inches. 

[Cygnus  olor,  J.  F.  Gmelin.    The  Mute  Swan. 

Favier  says  : — "  This  Swan  is  tolerably  numerous,  and  seen  flying  over 
near  Tangier  in  small  flights,  rarely  remaining  in  the  vicinity ;  but  they 
did  stay  in  1845  and  1849.  They  pass  south  in  Decemb3r,  returning  in 
April" 

Most  probably  the  above  refers  to  the  Whooper,  which  Favier  does  not 
mention;  as  it  is  unlikely  that  the  Mute  Swan  would  occur,  I  have  omitted  . 
to  number  the  species  in  my  list. 

Adults.  White  ;  bill  yellow,  with  black  tubercle  above  nostrils. 
Young.  Greyish  brown  ;  bill  lead-coloured.     Length  60  inches.] 

Subfamily  ANSERINE. 

230.   Anser  cinereus,  Meyer.    The  Grey-lag  Goose. 

/ 
Moorish.  Wiz.     Spanish.  Ganso,  Anser. 

The  above  names  equally  apply  to  A.  segetum. 

Favier's  notes  are  the  same  for  both  this  and  the  following 
species,  viz. : — "  This  Goose,  which  the  Arabs  confound  with 
Anser  segetum,  is  as  numerous  as  that  bird  near  Tangier,  arriving 
during  November  and  December.  They  retire  north  in  March, 
seldom  making  any  stay  near  Tangier ;  they  pass  on  to  the  large 
lakes  and  rivers." 

On  the  Spanish  side  of  the  Straits  the  Grey-lag  Goose  is 
found  in  some  winters  at  the  Laguna  de  la  Janda  and  in  the 
various  lagoons  of  the  marisma  of  the  Guadalquivir  in  enormous 
numbers.  They  generally  arrive  at  the  former  place  about  the  20th 
of  November,  the  earliest  that  I  noticed  in  two  consecutive  years 
being  on  the  8th  of  November  and  the  25th  of  October. 
Commencing  their  departure  about  the  14th  of  February,  they 
are  all  gone  by  the  first  week  in  March ;  Major  Verner  saw 


ANSER  ERYTHROrUS.  217 

many  on  the  9th  in  1889,  and  they  seem  for  the  most  part  to 
migrate  by  day.  Although,  like  Ducks,  they  "  flight "  at  night — 
as  a  rule,  rather  later  in  the  evening  and  later  in  the  morning — 
they  affect  particular  favourite  spots  and  pools  without  any  appa- 
rent reason  for  their  likes  and  dislikes,  some  places  never  being 
frequented  by  them. 

They  can  always  be  easily  distinguished  at  some  distance  on 
the  wing  by  the  ash-grey  of  the  shoulders,  which  colour,  when 
they  are  on  the  wing,  is  very  apparent,  and  they  always  can  be 
known  by  their  note  or  cry.  They  also,  when  flying,  make  a 
creaking  noise,  caused  by  the  stiff  primaries,  somewhat  resem- 
bling the  rattling  together  of  dry  reeds,  which  can  be  heard  only 
when  they  pass  very  close ;  and  a  very  joyous  sound  it  is  for  the 
gunner  to  hear. 

Rump  and  wing-coverts  asliy  grey r;  bill  flesh-colour;  nailte/<&  ;  legs  and 
feet  flesh-coloured. 

Adult.  White  round  base  of  bill ;  breast  with  a  few  black  marks. 
Length  35  inches. 

231.  Anser  albifrons,  Scopoli.    The  White-fronted  Goose. 
One  of  this  species  was  obtained  near  Tapatanilla  on  the  8th 

of  January,  1876,  by  Major  Kelham,  74th  Highlanders,  and 
another  was  got  near  Seville  by  Ruiz.  Lord  Lilford  further 
informed  me  that  he  recognized  the  cry  of  this  Goose  in  the 
marisma. 

Much  smaller  than  preceding  ;  bill  orange-yellow  ;  nail  white ;  legs  and 
feet  orange-yellow. 

Adults.  White  round  base  of  bill  ;  much  marked  with  black  on  breast, 
most  so  in  male. 

Young.  Without  black  on  breast,  and  in  first  plumage  without  white 
round  bill;  bill  and  feet  paler  than  in  adult,  and  nail  of  bill  brownish  black, 
pale  at  the  tip.  Length  27  inches. 

232.  Anser    erythropus,    Linnaeus.      Lesser   White-fronted 

Goose. 

A  male  of  this  small  race  was  obtained  near  Seville  in  February 


218  ANAT1D.E. 

1878,  and  I  clearly  made  out  two  among  a  lot  of  other  Geese  near 
the  Isla  Menor  in  March  1882. 

Resembles  the  last,  but  is  very  much  smaller,  and  the  markings 
of  adults  more  defined. 


233.  Anser  segetum  (J.  F.  Gmelin).    The  Bean-Goose. 

On  the  Spanish  side  of  the  Straits  this  species  is  much  less 
numerous  than  the  Grey-lag  Goose ;  and  it  was  some  time  before 
I  could  succeed  in  obtaining  a  specimen  for  identification.  As 
far  as  my  experience  goes,  I  should  say  the  present  species  occurs 
in  the  proportion  of  one  to  every  two  hundred  of  the  Grey-lag ; 
but  as  Favier  considers  both  kinds  equally  common  in  Morocco, 
perhaps  in  some  seasons  the  present  species  may  be  more  abun- 
dant than  in  others. 

Major  Verner  informs  me  (1894)  that  out  of  some  hundreds  of 
Geese  that  he  has  seen  killed  at  La  Janda  during  the  last  twenty 
years  he  only  once  saw  any  Bean-Geese,  when,  on  February  the 
loth,  1890,  he  shot  two  out  of  a  lot  of  seven. 

No  black  on  breast;  bill  blackish  brown,  nail  black;  legs  and  feet  orange- 
yellow.  Length  34  inches. 

The  Pink-footed  Goose  (Anser  brachyrhynchus,  Baillon),  which  does  not 
appear  to  have  yet  been  met  with  in  the  vicinity  of  the  Straits,  is  much  as 
last,  but  smaller  ;  nail  of  bill  black  ;  legs  and  feet  flesh-colour.  Length 
28  inches. 

234.  Bernicla  leucopsis,  Bechstein.    The  Bernicle  Goose. 

A  single  specimen  of  this  coast-frequenting  Goose,  obtained 
near  Seville  several  years  ago,  was  in  the  possession  of  the  land- 
lord of  the  Fonda  de  Europa,  possibly  an  escaped  bird  from  San 
Lucar,  and  is  the  only  instance  yet  known  of  its  occurrence  in 
Andalucia. 

Head,  neck,  and  throat  black  ;  front  and  sides  of  head  and  upper  throat 
white  ;  black  mark  between  eye  and  bill.  Length  25  inches. 


TADORNA  CASARCA.  219 


Subfamily  TADORNIN^E. 

235.  Tadorna  cornuta  (S.  G.  Gmelin).    The  Sheld-duck. 

Moorish.  Bou-he-baida  (Favier).  Spanish.  Pato  tarro, 
Anserata. 

"  This  species  is  not  regular  in  its  appearance  near  Tangier, 
and  occurs  between  November  and  February." — Favier. 

Not  observed  in  vicinity  of  Gibraltar,  this  coast-frequenting 
Sheld-duck  is  found  near  the  mouth  of  the  Guadalquivir,  where 
they  breed. 

Male.  Head  and  neck  green,  below  that  a  white  collar,  with  a  broad 
rich  chestnut  band,  covering  the  breast  j  wing-coverts  white  ;  wing-spot 
or  speculum,  formed  by  colours  of  the  outer  webs  of  the  secondaries, 
green  ;  fleshy  knob  on  base  of  upper  mandible  ;  bill  red  ;  legs  and  feet 
pink. 

Female.  Smaller  than  male,;  less  bright  markings,  and  no  knob  on  bill. 
Length  25-26  inches. 

236.  Tadorna  casarca  (Linnaeus).    The  Ruddy  Sheld-duck. 

Moorish.  Bou-he  (Favier}.  Spanish.  Pato  tarro,  Pato  canelo, 
Labanco. 

" This  species  is  resident  at  noc great  distance  from  Tangier; 
and  others  are  migratory,  crossing  to  Europe  during  April  and 
May,  returning  in  September  and  October.  In  the  immediate 
vicinity  of  Tangier  it  is  scarce  and  only  observed  in  small  lots  on 
the  lakes  and  large  rivers.  Often  they  entirely,  though  irregu- 
larly, disappear  for  months  at  a  time,  probably  going  to  marshes 
not  very  far  off.  The  months  during  which  they  are  usually 
absent  are  February,  March,  and  June." — Favier. 

They  were  repeatedly  to  be  seen  exposed  for  sale  in  Tangier 
market,  with  their  throats  slit  in  Mahometan  fashion,  but  are 
very  poor  food,  not  worth  cooking,  as  was  the  case  in  India, 
where  the  Ruddy  Sheld-duck  is  best  known  to  Anglo-Indians  as 
the  Brahminy  Duck. 


220  ANATID.E. 

They  are  quite  inland  birds,  though  occasionally  frequenting 
salt  lagoons,  and  are  much  more  frequent  on  the  African  than  on 
the  Spanish  side  of  the  Straits,  where  I  never  met  with  any, 
though  have  seen  a  few  in  Seville  market  in  spring  and  the 
end  of  autumn.  They  breed  near  the  mouth  of  the  Guadalquivir, 
and  used  to  do  so  at  Fuente  Piedra ;  also  no  doubt  elsewhere 
north  of  the  Straits.  They  nest  in  holes  in  cliffs  and  rocks ;  and 
an  account  is  given  of  one  nest  by  Mr.  Salvin,  in  '  The  Ibis  '  for 
1859  (p.  362).  In  Morocco  they  are  very  wary;  and  I  recollect, 
in  company  with  a  brother  officer,  trying  in  vain  to  approach 
three  which  we  saw  near  Vincent's  Farm,  Sharf  el.  Akab,  in 
October :  but  at  last  they  luckily  pitched  close  to  a  horse  feeding 
in  the  marsh ;  so  we  stalked  up  behind  the  animal  to  within 
twenty-five  yards  of  the  birds,  bagging  all  three. 

General  colour  rufous  buff;  crown  light  buff;  wing-coverts  white  ; 
wing-spot  green  ;  bill,  legs,  and  feet  blackish. 

Sexes  alike,  except  that  the  male  has  a  blackish  ring  round  neck.  Length 
25  inches. 


Subfamily  ANATIN^I. 

237.  Anas   boscas,    Linnaeus.      The   Wild   Duck    (Mallard, 
male). 

Moorish.  Zerak  el  ras  (Blue  head).     Spanish.  Pato  real. 

"  The  Wild  Duck  frequents  the  vicinity  of  Tangier  throughout 
the  year.  Those  which  are  not  resident  cross  to  Europe  during 
March  and  April,  returning  in  November  and  December.  Those 
which  remain  to  nest  begin  to  lay  during  the  month  of  February  ; 
and  eggs  may  be  occasionally  found  as  late  as  the  beginning  of 
June.  Tame  Ducks  are  called  Bourk  by  the  Arabs." — Favier. 

The  Moorish  name  given  above  is  not  appropriate,  the  head  of 
the  Mallard  being  green. 

In  Andalucia  the  present  species  is  abundant  in  winter ;  and  a 
considerable  number  remain  for  the  breeding-season,  hatching  by 


ANAS  STREPERA.  221 

about  the  25th  of  April ;  but  are  so  molested  by  egging  that  it 
is  a  wonder  that  any  young  are  brought  up. 

Wing-spot  purple  ;  tail  of  sixteen  feathers  ;  bill  rather  longer  than 
head. 

Male.  Head  and  neck  green  ;  four  central  tail-coverts  black,  lengthened 
and  curled  upwards. 

Female.  General  colour  brown. 

Male  in  summer.  Resembles  female.     Length  21  inches. 

238.  Anas  strepera,  Linnaeus.    The  Gadwall. 

"  This  species  is  as  scarce  near  Tangier  as  Fuligula  rujina, 
the  -  Red-crested  Pochard  ;  and  their  appearance,  which  takes 
place  between  February  and  March,  is  irregular  and  uncertain." 
— Favier. 

The  Gadwall,  on  the  Spanish  side  of  the  Straits,  cannot  be 
termed  a  common  bird.  I  only  met  with  them  on  a  few 
occasions :  one  shot  at  flight  on  the  26th  of  November,  1869, 
another  shot  at  flight  on  the  22nd  of  December,  1871,  at 
Tapatanilla,  and  three  others  killed  there  in  February  1874,  are 
my  only  personal  experiences  ;  but  I  have  seen  them  in  Seville 
market  in  February  and  March. 

Lord  Lilford  informed  me  that  he  saw  ten  or  twelve  Gadwall 
at  the  lakes  of  Santa  Olaya,  in  the  Goto  de  Donana,  in  the  early 
part  of  May,  and  considered  that  they  were  breeding,  although 
he  was  unable  to  discover  a  nest  then;  afterwards  they  were 
found  breeding  there.  He  also  informed  me  that  the  local  name 
there  was  "  Frisa " — a  word  which  signifies  coarse  cloth  or 
frieze,  probably  from  grey  colour.  The  Portuguese  name  given 
by  Arevalo  for  this  Duck  is  "  Frisada." 

The  note  of  the  Gadwall,  in  winter,  is  a  hoarse  croak,  by 
which  they  can  be  distinguished  at  "  flight." 

Wing-spot  white  ;  bill  shorter  than  head,  and  narrowing  slightly  towards 
tip.  General  colour  grey. 

Male.  Chestnut  on  shoulder  of  wing. 
Female.  Brown  and  grey.     Length  20  inches. 


222  ANAT1D.E. 

239.  Spatula  olypeata  (Linnaeus).    The  Shoveler. 
Moorish.    Bou-zlafa  (father  of  the  bowl).     Spanish.  Paleton, 

Sardinero,  Pato  cuchareta. 

"  This  Duck  is,  in  some  winters,  common  near  Tangier,  arriving 
during  September  and  October,  leaving  for  the  north  in  February 
and  March." — Favier. 

The  Shoveler  is  met  with  in  considerable  numbers  on  the 
Spanish  side  of  the  Straits,  and  chiefly  frequents  shallow  waters, 
as  they  do  in  every  country  in  which  I  "have  seen  them.  They 
mostly  arrive  around  Gasas  Viejas  and  on  the  Laguna  de  Janda  in 
October ;  and  I  have  known  of  their  occurrence  there  in  August, 
but  have  no  certain  knowledge  of  their  nesting  in  the  neigh- 
bourhood ;  indeed  I  never  saw  any  later  than  the  end  of  April, 
but  Lord  Lilford  informs  me  he  has  fairly  good  evidence  of  their 
breeding  in  the  marisma  of  the  Guadalquivir. 

The  only  species  of  the  genus  found  in  Europe ;  owing  to  the 
shape  of  the  bill,  they  cannot  be  well  mistaken  for  any  other 
Duck. 

Bill  broadly  dilated  at  the  end,  being  twice  as  broad  near  the  tip  as  at 
the  base  ;  wing-spot  green. 

Male.  Head  and  neck  green  ;  shoulders  pale  blue  ;  iris  yellow. 

Female.  Dark  brown. 

Male  in  summer.  Resembles  the  female.     Length  20  inches. 

240.  Daflla  acuta  (Linnaeus).    The  Pintail. 

Moorish.  Bou-sbula^y^r)  ("sbula"  is  "ear  of  corn";  "seboula," 
a  "  two-edged  dagger  ").  Spanish.  Pato  rabiido,  Pato  careto. 

"This  Duck  is,  during  the  winter  season,  nearly  as  plentiful 
in  the  vicinity  of  Tangier  as  the  Wild  Duck  (Anas  boschas). 
They  arrive  during  September  and  October,  and  leave  in  April 
and  May." — Favier. 

On  the  Spanish  side  I  never  found  the  Pintail  before  the 
month  of  November ;  it  is  exceedingly  abundant  in  some  winters 
on  the  Laguna  de  la  Janda  and  other  large  open  pieces  of  water, 


MARMAROXETTA  ANGUS!  IROSTRIS.  223 

and  is  in  consequence  very  difficult  to  shoot  by  day  without  the 
aid  of  a  stalking-horse.  These  Ducks  mostly  depart  during 
March,  some  lingering  on  later  into  the  month  of  April,  the 
latest  that  I  myself  observed  being  on  the  uth  of  that  month. 

Tail  graduated,  of  sixteen  feathers,  two  centre  feathers  much  elongated 
in  male,  slightly  so  in  female  ;  wing-spot  dark  green,  glossed  with  copper  ; 
bill  as  long  as  head. 

Male.  Head,  cheeks,  chin,  sides  of  neck,  and  upper  neck  in  front  rich 
dark  brown  ;  two  centre  tail-feathers  black. 

Female.  Brown  above,  greyish  white  below.     Length  24-28  inches. 

241.  Marmaronetta  ang-ustirostris  (Menetries).    The  Marbled 
Duck. 

Moorish.  Ashab  (grey),  Bourk  el  biad,  (white  duck)  (D.  Hay). 
Andaludan.  Ruhilla,  Pardilla. 

This  Duck,  on  both  sides  of  the  Straits, -appears  in  spring,  to 
remain  only  for  the  breeding-season,  and  is  exceedingly  abundant 
in  Morocco,  where,  at  the  lakes  of  Ras  el  Doura  in  April  1871, 
we  saw  flocks  numbering  many  hundreds  ;  and  they  are  frequently 
seen  exposed  for  sale  in  Tangier  market. 

Favier  states  that  they  arrive  during  March  and  April,- 
departing  in  October,  and  that  after  the  Common  Teal  they  rank 
as  the  most  common  Duck  in  the  country. 

On  the  Spanish  side  I  heard  of  three  being  seen  at  the  end  of 
February,  and  saw  six  or  seven  myself  on  the  23rd  of  March  ; 
but  the  majority  do  not  appear  until  late  in  April,  though  I  have 
noticed  them  passing  near  Gibraltar  early  in  that  month.  As  a 
rule,  they  mostly  leave  by  the  end  of  September ;  but  of  course 
stragglers  remain  later. 

The  Marbled  Duck  breeds  during  the  last  week  in  May, 
nesting  in  patches  of  rushes.  The  nest  is  like  that  of  a  Teal, 
containing  a  good  deal  of  the  down  from  the  breast  of  the  female ; 
and  eleven  eggs  appear  to  be  the  usual  complement.  The  latter 
much  resemble  those  of  the  Common  Teal,  being  of  a  yellowish- 
white  colour.  Favier  states  that  they  also  nest  in  rushes  during 


224  ANATIDJ3. 

May  and  June,  and  that  incubation  lasts  from  twenty-five  to 
twenty-seven  days. 

I  was  unable  to  find  this  Duck  near  Casas  Yiejas  or  about  the 
Laguna  de  la  Janda  ;  but  in  the  marisma  of  the  Guadalquivir, 
especially  near  the  Goto  del  Eey,  they  are  very  numerous. 

In  flight,  the  Marbled  Duck  somewhat  resembles  the  female 
Pintail  ;  but  it  is  more  of  a  Teal,  as  Lord  Lilford  observes.  We 
found  them  in  Morocco  wary  and  difficult  to  approach  ;  but 
in  the  dusk  they  "  flight  "  very  low,  and  by  watching  the  direction 
taken  by  them  for  one  night  you  may  on  the  next  evening  be 
tolerably  certain  of  shooting  a  good  many  ;  and  they  are  excellent 
eating. 

Sexes  alike.     Bill  narrow. 

Entire   plumage   light  brown   and  grey.     Secondaries  creamy  brown. 
Bill,  legs,  and  feet  black.     Length  14£  inches. 


242.  Querquedula  crecca  (Linnseus).    The  Teal. 

Moorish.  Frifar  (Favier).     Spanish.  Sarceta. 

"This  species  is  abundant  near  Tangier  during  winter,  passing 
north  in  February  and  March,  returning  in  September  and 
October."  —  Favier. 

The  Teal  is  very  numerous  on  the  Spanish  side,  and,  from 
their  habit  of  haunting  small  streams  and  marshes,  more  easily 
shot  than  any  other  of  the  Ducks.  Their  numbers  vary  con- 
siderably, as  in  some  seasons  they  are  much  more  abundant  than 
in  others.  They  chiefly  arrive  during  October,  leaving  for  the 
north  in  March,  and  have  been  known,  though  very  rarely,  to 
breed  near  Casas  Viejas.  Lord  Lilford  observed  a  single  bird  at 
Santa  Olaya  in  May  1872,  and  has  fairly  good  evidence  of  their 
nesting  in  the  district  since  that  date. 

Wing-spot  black  and  green  ;  tail  of  sixteen  feathers. 

Male.  Head  and  neck  chestnut,  with  broad  green  stripe  from  eye  down 
sides  of  neck,  margined  on  sides  of  head  with  buff. 

Female.  Above  dark  brown,  feathers  edged  with  grey  ;  below  whitish. 
Length  13-15  inches. 


MARECA  PENELOPE.  225 

243.  Querquedula    circia    (Linnaeus).       The  'Garganey    or 

Summer  Teal. 

Spanish.  Capitan. 

"  This  Teal  appears  irregularly  near  Tangier,  only  on  migration, 
and  does  not  occur  every  year.  They  arrive  during  February  and 
March,  passing  on  to  the  north,  and  are  seen  returning  south  in 
September." — Favier. 

The  Garganey  seems  to  be  equally  irregular  in  its  appearance 
in  Andalucia,  as  I  only  saw  one  in  Seville  market  in  March 
1869,  and  did  not  again  meet  with  any  till  March  1874, 
when  they  were  for  a  few  days  not  uncommon  about  Casas 
Viejas  and  near  Seville.  Lord  Lilford  informs  me  that  he  saw  a 
pair  at  Santa  Olaya,  near  Rocio,  in  May  1872,  and  that  since 
then  they  have  been  found  nesting  there. 

Male.  Wing-spot  green,  between  two  bars  of  white  ;  wing-coverts  or 
shoulders  bluish  grey  ;  neck  and  breast  dark  brown,  with  pale  brown 
crescent-shaped  bands. 

Female.  Much  as  female  of  Q.  crecca,  but  without  green  metallic 
speculum  ;  shoulders  with  very  slight  grey  tinge.  Length  15-16  inches. 

244.  Mareca  penelope  (Linnaeus).    The  Wig-eon. 

Moorish.  Bou-khe-siwa.  Spanish.  Silbon  (Whistler),  Pato 
franciscano. 

"  This  species  is  the  most  abundant  Duck  near  Tangier,  being 
found  in  large  flocks  throughout  the  winter  months.  They 
commence  to  arrive  in  August  and  September,  and  leave  during 
March  and  April." — Favier. 

Exactly  the  same  may  be  said  of  the  Wigeon  on  the  Spanish 
side  of  the  Straits,  except  that  I  never  saw  any  so  early  as  Favier 
mentions.  They  begin  to  arrive  early  in  October ;  but  the 
greater  number  do  not  appear  until  November;  and  they  are 
then  by  far  the  most  common  Duck,  in  some  winters  swarming 
in  thousands  on  the  Laguna  de  la  Janda.  Their  departure  for 
the  north  begins  about  the  end  of  March  ;  but  a  few  linger  on 
throughout  the  whole  of  April. 

Q 


226  AXATID.E. 

Tail  of  fourteen  feathers  ;  bill  shorter  than  head. 

Male.  Forehead  and  crown  huffish  white  ;  wing-coverts  white  ;  back 
white,  finely  vermiculated  with  black  ;  wing-spot  green. 

Female.  Above  brown  and  greyish  brown  ;  head  and  neck  mottled  with 
light  brown  and  blackish  brown  ;  shoulders  of  wings  greyish  white  ;  wing- 
spot  greyish. 

Males  in  summer  lose  the  whitish  crown,  and  to  a  certain  extent  resemble 
the  females.  Length  20  inches. 


Subfamily  FULIGULIN.E. 

245.  Fuligula  ruflla  (Pallas).    The  Red-crested  Pochard. 

"  This  Duck  is  accidentally  met  with  around  Tangier,  but  is  a- 
very  rare  species.  I  only  obtained  two — one  in  1835,  the  other 
in  1849."— Favier. 

I  never  met  with  this  Pochard  on  either  side  of  the  Straits, 
and  have  seen  but  one  specimen  said  to  be  Andalucian.  A  more 
eastern  species  (frequenting  still,  deep  waters,  and  seen  rarely  on 
rivers),  they  are  of  more  common  occurrence  in  the  south-east  of 
Spain;  and  Lord  Lilford  mentions  them  as  common  on  the 
Albufera  near  Valencia,  where  they  used  to  breed. 

Wing-spot  white  ;  tail  of  fourteen  feathers,  as  in  next  four  species. 

Male.  Head  and  upper  neck  reddish  chestnut',  feathers  on  crown  forming 
an  erectile  crest  ;  back  yellowish  brown  ;  lower  neck,  breast,  and  belly 
black  ;  bill  and  legs  vermilion  ;  iris  red. 

Female.  Crown  dark  brown  without  crest ;  cheeks,  neck,  and  sides  of 
throat  light  grey  ;  bill  and  legs  reddish  brown.  Length  21  inches. 

246.  Fuligrula  ferina  (Linnaeus).    The  Common  Pochard. 

Spanish.  Cabezon,  Cenizo. 

"This  species  arrives  during  October  to  remain  in  Morocco 
for  the  winter,  departing  for  the  north  in  April  and  May." — 
Favier. 

We  found  the  Pochard  common  about  the  lakes  near  Tetuan, 
and  shot  one  there  as  late  as  the  30th  of  March.  On  the  Spanish 
side  of  the  Straits  I  have  rarely  seen  this  Duck  near  Gibraltar, 


FUL1GULA  NYROCA.  227 

and  then  only  in  winter;  but  at  times  a  good  many  are  to 
be  seen  at  the  Laguna  de  la  Janda.  In  the  marisma  below 
Seville  they  are  much  more  abundant,  and  Lord  Lilford  informs 
me  "that  in  August  1892  a  female  Pochard  and  a  quantity  of 
eggs  were  brought  to  him  from  the  marisma ;  and  the  man  who 
brought  them  declared  that  in  the  previous  spring  and  summer, 
Pochards,  which  were  almost  unknown  there,  were  only  slightly 
less  abundant  than  Marbled  Ducks,  which  latter  were  in 
most  extraordinary  profusion  owing  to  the  abundance  of 
water." 

In  the  spring  of  that  year,  1892,  I  never  saw  the  country 
around  Gibraltar  so  flooded,  the  Laguna  de  la  Janda  resembling 
an  inland  sea. 

"Wing-spot  grey  ;  bill  black,  longer  than  head,  broad  band  of  blue  across 
the  middle. 

Male.  Head  and  neck  chestnut-red  ;  breast  and  upper  back  black  ;  back 
and  scapulars  white,  finely  vermiculated  with  black  ;  iris  red. 

Female.  Markings  as  in  male,  but  with  dull  brown  head  and  neck  ;  dark 
brown  breast  and  upper  back.  Length  17—19  inches. 

247.   Fuligula    nyroca    (Guldenstadt).        The    White  -  eyed 
Pochard  or  Ferruginous  Duck. 

Moorish.  Ziriguil  (Favier).     Andalucian.  Negrete,  Pardote. 

The  White-eyed  Pochard  may  be  considered,  like  the  Marbled 
Duck,  a  summer  resident  on  both  sides  of  the  Straits,  and  is  most 
abundant  in  Morocco. 

I  saw  many  hundreds  at  the  lakes  of  Has  el  Doura  towards  the 
end  of  April,  being  even  then  in  large  flocks.  We  shot  them  at 
flight  in  the  evening  at  the  same  time  as  some  Marbled  Ducks  ; 
but  the  two  species  did  not  fly  together.  There  were  also  a  few 
of  the  White-eyed  Pochards  about  the  lake  of  Esmir  at  the  end 
of  March.  Favier  writes  of  the  present  species,  that  it  is 
"  abundant  near  Tangier,  arriving  from  the  south  during  May 
and  departing  in  November  and  December,  totally  disappearing 

Q2 


228  ANATID.E. 

for  a  time  in  winter  *.  They  are  most  abundant  at  Ras  el  Doura, 
breeding  in  June  and  July,  the  incubation  lasting  thirty  days." 

On  the  Spanish  side  this  Duck  is  common  during  the  breeding- 
season  in  some  parts  of  the  marismas,  and  commences  to  nest 
about  the  end  of  April.  Lord  Lilford  obtained  a  nest  in  May 
1872,  in  the  Goto  de  Donana,  composed  of  dead  dry  water-plants, 
flags,  &c.,  lined  with  thick  brownish- white  down  and  a  few  white 
feathers.  It  was  placed  at  a  short  distance  from  the  water,  in 
high  rushes,  and  contained  nine  eggs.  Although  they  generally 
pass  south  early  in  autumn,  some  are  found  occasionally 
in  the  winter  months,  I  once  saw  and  shot  a  single  bird  as 
late  as  the  6th  of  December,  which,  albeit  in  fair  condition,  from 
its  excessive  tameness,  was  probably  from  some  cause  incapable 
of  migration. 

I  have  always  found  this  Duck,  like  its  allies,  F.  rufina  and 
F.ferina,  frequenting  deep,  still,  weedy  water  rather  than  shallow 
open  places  ;  and  the  flesh  of  the  present  species  is  not  only,  like 
theirs,  excellent  eating,  but  far  surpasses  either  in  that  respect. 

Wing-spot  wldte  ;  tail  short,  of  fourteen  feathers,  much  graduated  ;  bill 
lead-coloured,  longer  than  head. 

Male.  Head,  neck,  and  general  colour  above  chestnut-brown  ;  under- 
parts  greyish  white  ;  iris  wldte. 

Female.  Marked  as  male,  but  duller  ;  iris  dull  white.  Length  15-17 
inches. 

248.  Fuligula  cristata  (Leach).    The  Tufted  Duck. 

"  This  species  is  in  some  years  very  abundant  near  Tangier, 
arriving  here  for  the  winter  in  November,  and  returning  north 
during  February.  In  some  seasons  they  are  not  to  be  met  with, 
but  were  common  in  the  years  1845,  1846,  1849,  1850,  1858, 
and  1861."— Favier. 

Kelham  saw  great  numbers  at  Esmir  in  March  1881. 

*  It  is  remarkable  that  the  instances  recorded  of  this  Duck's  appearance  in 
the  British  Islands  should  have  been  often  iu  the  winter  months. 


(EDEMIA  NIGItA.  229 

The  Tufted  Duck  is  sometimes  plentiful  in  winter  on  the 
Laguna  de  la  Janda,  is  well  known  in  the  marisma,  and  I  have 
occasionally  seen  them  in  the  Bay  of  Gibraltar. 

Wing-spot  white. 

Male.  Head  and  neck  glossy  purplish  black,  with  an  occipital  crest  or 
tuft ;  breast,  belly,  sides,  flanks,  and  speculum  white  ;  the  rest  of  plumage 
black  ;  iris  yellow. 

Female.  Crest  smaller  ;  brown  where  black  in  the  male  ;  underparts 
brownish  white.  Length  17  inches. 

249.  Fuligula  marila  (Linnaeus).    The  Scaup  Duck. 

This  Duck  is  a  rare  visitant  in  the  Straits,  but  has  occurred  in 
the  Bay  of  Gibraltar  in  December  ;  marine  and  coast-frequenting, 
they  are  not  liable  to  much  notice  and  seldom  likely  to  appear  in 
the  markets. 

Wing-spot  white  ;  bill  very  broad,  about  as  long  as  head. 

Male.  Head,  neck,  upper  breast,  rump,  and  under  tail-coverts  black, 
with  green  gloss  ;  back  and  scapulars  white,  broadly  verrniculated  with 
black  ;  belly  white  ;  iris  straw-yellow. 

Female  and  young  male.  White  band  round  base  of  bill,  broadest  in, 
female  ;  head  and  neck  dark  brown,  as  are  all  the  parts  which  are  black  iu 
the  male.  Length  18  inches. 

250.  Clangula  glaucion  (Linnaeus).    The  Golden-eye. 

All  to  be  stated  regarding  this  northern  species  is  that  they 
rarely  occur  about  the  Straits  in  winter. 

Bill  much  shorter  than  head  ;  tail  of  sixteen  feathers,  as  in  next. 

Male.  Head  and  neck  glossy  green  ;  feathers  on  crown  slightly 
lengthened  ;  small  ivhite  spot  at  base  of  bill  ;  scapulars  white  ;  wing-spot 
white. 

Female  and  young  male.  Head  and  neck  ash-brown  ;  no  white  spot  on 
head  ;  wing-spot  white,  divided  by  a  black  line.  Length  16-11)  inches. 

251.  CEdemia   nigra    (Linnaeus).      The    Common    Scoter   or 

Black  Duck. 

Moorish.  Bourk  -  el  -  behar  (Sea  -  Duck).  Spanish,  Malaga. 
Coquinero,  Pato  negro. 


230  A\ATID.E. 

Favier  states  that  the  Scoter  is  "  found  in  abundance  near 
Tangier,  arriving  sometimes  as  early  as  August,  retiring  north- 
wards in  April." 

I  found  this  Duck  in  some  seasons  very  common  about  the 
Straits,  especially  after  rough  weather  in  Gibraltar  Bay;  but 
they  do  not  appear  except  in  small  lots.  The  earliest  noticed 
was  on  the  12th  of  November ;  the  latest  on  the  12th  of  March, 
1872  ;  and  was  very  abundant  off  Malaga  in  March  1882. 

Adult  male.  Entirely  black;  bill  black,  with  centre  of  upper  mandible 
yellow,  and  a  knob  at  the  base. 

Female.  Sooty  brown  ;  chin  and  throat  whitish ;  side  of  head  greyish 
•white  ;  bill  without  knob. 

Young.  Cheeks,  chin,  sides,  and  front  of  neck  dull  greyish  white;  belly 
speckled  with  brown  and  white.  Length  21  inches. 

[CEdemia  fusca  (Linnaeus).    Trie  Velvet  Scoter. 

Is  recorded  from  Valencia,  by  Vidal,  but  has  not  been  observed 
within  the  district  treated  of  in  this  work, 

Adult  male.  Velvet-black,  except  the  large  wing-spot,  eyelid,  and  patch 
below  eye,  which  are  white  ;  iris  yellowish  white. 

Female.  Blackish  brown,  but  with  white  wing-spot.     Length  21  inches.] 


Subfamily  ERISMATURIN^. 

252.  Erismatura  leucocephala  (Scopoli).     The   Spiny-tailed 
or  White-faced  Duck. 

Spanish.  Porron. 

"  This  species  occurs  near  Tangier  on  passage,  passing  north 
during  April  and  returning  to  winter  further  south  in  October. 
Some  of  the  spring  migrants  remain  in  the  country  to  breed  in 
June,  laying  as  many  as  ten  pure  white  eggs,  with  a  rough 
granulated  surface.  This  Duck  is  not  at  all  regular  in  its 
appearance,  but  in  some  seasons  is  quite  common." — Favier. 

In  Andalucia  the  Spiny-tailed  Duck  occurs  as  above,   chiefly 


MERGUS  SERRATOR.  231 

on  the  coast  and  on  large  lakes,  but  T  never  met  with  any  near 
Gibraltar. 

Bill  longer  than  head,  much  swollen  at  the  base. 

&  / 

Tail  long,  pointed,  and  stiff. 

General  colour  chestnut  to  buff,  vermiculated  with  blackish  grey. 

Male.  Crown  black  ;  forehead,  sides  of  head,  chin,  and  nape  white. 

Female.  Crown  and  nape  dark  brown.  General  colour  more  chestnut 
than  in  male,  white  streak  below  eye  to  nape,  chin  and  upper  throat  white. 
Length  17^  inches. 

Subfamily  MERGING. 

The  following — Mergansers,  or  "  Saw-bills  " — have  the  bill 
slender  and  straight,  the  edges  with  strong  saw-like  teeth 
pointing  backwards. 

253.  Mergris  merganser,  Linnseus.    The  Goosander. 

The  Goosander  is  recorded  by  Favier  as  having  been  once 
obtained  by  him  near  Tangier  in  October  1862,  and  another  was 
found  dead  on  the  shore  near  that  town  during  the  winter  of 
1869-70,  the  only  instance  in  which  I  met  with  any.  Arevalo 
records  this  species  from  Malaga. 

The  immature  males  of  this  and  the  next  two  species  can  be 
at  once  distinguished  by  the  trachea,  which  in  the  males  has 
more  or  less  bony  enlargement  at  the  base,  while  in  the  females 
it  is  uniform  in  size  throughout  the  entire  length. 

Male.  Head  and  upper  neck  glossy  metallic  black,  slightly  crested  ; 
lower  neck  and  underparts  white ;  rump  and  tail  ashy  grey  ;  bill  blood-red, 
and,  as  in  next,  longer  than  head  ;  iris  red. 

Female.  Head  and  upper  neck  light  chestnut,  crested ;  above  slate-grey  ; 
below  white  ;  wing-spot  white. 

Young.  Like  female,  but  without  crest.     Length  26  inches. 

254.  Mergns   serrator,   Linnseus.       The   Red-breasted    Mer- 

g-anser. 

Spanish.  Pato  cle  sierra  (Saw-duck). 

This  species  is  not  mentioned  in  Favier's  MS.,  but  is  found  in 


232  COLUMBID.E. 

some  winters  in  considerable  numbers  in  the  Bay  of  Gibraltar, 
generally  during  December  and  January. 

Male.  Head  and  neck  glossy  metallic  black  ;  crest  of  narrow  elongated 
black  feathers ;  white  collar  on  neck,  showing  much  when  flying  ;  rump 
white,  with  wavy  black  lines  ;  tail  uniform  brown  ;  bill  vermilion  ;  legs 
orange  ;  iris  red. 

Female  and  young.  Like  those  of  preceding  species,  but  much  smaller, 
and  head  darker,  black  bar  across  wing-spot.  Length  22-24  inches. 

255.  Mergus  albellus,  Linnaeus.    The  Smew. 
The  Smew  occurs  in  some  winters  about  the  Straits  in  imma- 
ture plumage. 

Bill  shorter  than  head. 

Male.  Black,  grey,  and  white.  Head  and  neck  white ;  black  stripe 
through  eye  to  nape;  slightly  crested  ;  ramp  ash-grey. 

Female.  Head  and  nape  chestnut,  crested  ;  above  slate-grey  ;  black  patch 
from  bill  to  eye  ;  rump  greyish  black. 

Young  male.  Like  female,  but  has  no  black  patch  between  eye  and  bill. 
Length  14-17  inches. 


Order  COLTJMB^.         Family  COLUMBIA. 

256.  Columba  palumbus,  Linnaeus.    The  Ring-Dove  or  Wood- 
Pigeon. 

Moorish.  Kamor.     Spanish.  Paloma  torcaz. 

"This  Pigeon  is  found  near  Tangier  throughout  the  year. 
Some  are  migratory,  crossing  to  Europe  in  March  and  April." — 
Favier. 

In  some  localities  in  Morocco  the  Wood-Pigeon  positively 
swarms.  In  April,  up  a  valley  near  the  Fondak,  to  the  south- 
west of  that  place,  on  the  road  between  Tangier  and  Tetuan,  it 
would  have  been  easy  to  shoot  a  hundred  in  a  day,  they  were  in 
such  numbers  and  so  excessively  tame.  Two  or  three,  which  we 
shot  to  eat,  had  their  crops  full  of  the  tuberous  root  of  some 
weed  which  had  been  ploughed  up  and  was  lying  in  quantities 
about  the  fallow  fields.  During  the  same  month,  about  three 


COLUMBA  (EN AS.  233 

years  previously,  we  noticed  considerable  numbers  near  Larache  ; 
but  there  they  were  much  more  wild,  though  not  so  shy  as  in 
England  or  Andalucia.  In  the  latter  country,  about  Gibraltar, 
a  few  pairs  nest  in  the  Cork-wood  and  other  wooded  districts ; 
they  are  most  abundant  during  the  winter  months,  though  I 
never  saw  any  great  quantity. 

Adult.   White  on  outer  wing-coverts  ;  white  patch  on  each  side  of  neck. 
Young.  Without  white  patch  on  neck.     Length  16-17  inches. 

257.  Columba  oenas,  Linnaeus.    The  Stock-Dove. 

Moorish.  Hamam  el  Berri  (Wild  Pigeon). 

The  Stock-Dove  is  neither  mentioned  by  Favier  nor  Mr.  Drake 
as  occurring  in  Morocco.  We  found  them  near  the  Fondak 
between  Tangier  and  Tetuan  at  the  same  time  and  place  that  the 
Bing-Doves  were  so  abundant,  and  they  are  sufficiently  common 
to  be  known  to  the  Moors  there  by  the  above-mentioned  name, 
which,  by  the  way,  is  the  same  as  that  used  for  the  next  species, 
C.  lima.  They  were  in  some  numbers,  and  we  shot  one  or  two 
for  identification,  being  further  informed  by  the  Moors  that  they 
nested  in  holes  of  trees.  They  evidently  were  breeding  at  that 
time  ;  but  we  failed  to  discover  a  nest  during  the  Tery  short 
period  that  we  remained  there.  We  also  noticed  the  Stock-Dove 
in  April  near  Larache,  and  Meade-Waldo  observed  them  in 
March  1894. 

On  the  Spanish  side  of  the  Straits  I  only  observed  this  species 
once  near  Gibraltar,  but  Arevalo  mentions  them  as  -  occurring 
near  Malaga. 

The  absence  of  the  white  patches  on  the  carpal  joints  of  the 
wings  will  often  serve  to  distinguish  them,  when  flying,  from  the 
King-Dove,  independently  of  their  smaller  size,  while  the  absence 
of  the  white  rump  equally  distinguishes  them  from  the  Rock- 
Dove — not  that  the  latter  species  is  usually  met  with  in  the  same 
locality.  .  .  . 

Adult.  Sides  of  neck  with  metallic-green  patch  ;  axillaries  grey. 
Young.  Without  patch  on  neck.     Length  13  inches. 


234  COLUMBIA. 

258.  Columba  livia,  J.  F.  Gmelin.    The  Rock-Dove. 
Moorish.  Ham  am  el  Berri.     Spanish.  Zurita,  Paloma  brava. 
"This  is  the  most  common  of  the  Pigeons  about   Tangier, 

living  in  rocks  and  even  in  the  ramparts  of  the  town,  breeding 
both  in  a  wild  and  in  a  domestic  state  all  the  year  round." — 
Favier. 

The  Rock-Dove  is  plentiful  on  both  sides  of  the  Straits 
wherever  there  are  rocks  and  caves,  inland  as  well  as  on  the 
coast.  Many  are  resident  at  Gibraltar,  on  North  Front  and  at 
the  "  back  of  the  Rock  " ;  and  at  one  time  some  sport  was  to  be 
obtained  upon  getting  permission  to  shoot  them  ;  but  a  "  young 
and  inexperienced  "  arrival  one  day,  instead  of  killing  pigeons, 
shot  one  or  more  of  the  celebrated  Gibraltar  Apes,  for  which  he, 
amongst  suffering  other  indignities  and  punishments,  was  after- 
wards known  as  "  Du  Chaillu."  In  consequence  of  this  exploit, 
all  leave  to  shoot  there  was  thenceforth  withheld,  and  very 
rightly  so. 

Pied  and  white  varieties  are  frequently  seen  ;  whether  escaped 
tame  Pigeons  or  real  wild  Rock-Doves,  I  cannot  say,  probably 
the  former. 

Lower  back  and  rump  white  ;  two  conspicuous  black  bands  across  the 
wings  ;  axillaries  white.  Length  13  inches. 

259.  Turtur  communis,  Selby.    The  Common  Turtle-Dove. 
Moorish.  Imam,  Stitsia,  Targal.     /Spanish.  Tortola. 

"  Is  a  summer  resident  near  Tangier,  vast  numbers  arriving  to 
cross  the  Straits  in  flocks  during  April  and  May,  returning  in 
September  and  October,  then  to  retire  south  for  the  winter. 
This  species  is  without  doubt  the  origin  of  the  domestic  Turtle- 
Dove  *,  called  limama  or  deJcrallah  (praise  of  God),  some  of 
which  birds  are  pure  white." — Favier. 

*  Favier's  opinion  is  given,  but  it  is  not  certain  which  species  is  the  origin  of 
the  domestic  Dove. 


TURTUR  SEXEGALENSIS.  235 

The  Common  Turtle-Dove  is  seen  in  extreme  abundance  in 
Andalucia,  during  its  stay  being  a  great  object  of  pursuit  to  the 
Spanish  tirador,  who,  in  August,  often  makes  a  puesto,  or  hiding- 
place,  near  some  favourite  drinking-haunt  of  the  doves,  and 
shoots  them  much  in  the  same  way  as  the  Partridge  —  that  is,  on 
the  ground,  three  or  four  in  a  row  ;  only  with  the  Doves  he  has, 
of  course,  no  reclame,  or  call-bird. 

The  Turtle-Dove  chiefly  arrives  during  the  first  week  in  May, 
more  coming  in  that  week  than  during  all  the  rest  of  their 
migration,  which  is  about  the  time  that  the  seed  of  the  Greater 
Honey  wort  (Cerinthe  major]  is  ripe,  and  on  which  they  feed  on 
arrival.  I  first  saw  one  in  1870  on  the  llth,  and  in  1872  on  the 
14th  of  April.  They  mostly  disappear  by  the  beginning  of 
October.  The  latest  I  noticed  was  a  single  bird  at  Casas  Viejas 
on  the  31st  of  October.  In  my  note-book  I  have  one  recorded 
as  seen  on  the  9th  of  October,  in  the  middle  of  the  Bay  of 
Biscay. 

Head,  nape,  rump,  and  outer  wing-coverts  bluish  ash  ;  under  tail-coverts 
white  ;  collar-patch  of  four  rows  of  black  feathers  tipped  with  white  on 
each  side  of  neck;  tail  graduated,  two  centre  feathers  brown,  the  rest 
slate-grey,  the  ends  and  outside  web  of  outer  pair  white. 

Young.  Without  neck-patch.     Length  11^  inches. 


260.  Turtur  senegralensis  (Linnaeus).    The  Egyptian  Turtle- 
Dove. 

This  species  is  mentioned  by  Mr.  Drake  as  found  commonly 
in  the  southern  part  of  Morocco,  but  does  not  appear  to  have 
been  obtained  by  Favier,  nor  were  any  met  with  by  us,  in  the 
north-west  of  that  country. 

The  absence  of  spots  on  the  back,  and  its  smaller  size,  will  distinguish  it 
from  any  other  species  likely  to  be  met  with  in  Morocco. 

Feathers  of  the  fore  neck  bifurcated,  black,  with  two  rufous  spots  at  end 
of  each,  forming  collar  ;  rump  bluish  lead-colour.  Length  1  1  inches. 


236  PTEROCLID.E. 


Order  PTEROCLETES.         Family  PTEROCLID^. 

261.  Pteroclurus  alchata  (Linnaeus).     The  Pin-tailed  Sand- 
Grouse. 

Moorish.  El  Gata.     Spanish.  Ganga. 

Favier  states  this  species  to  be  "  scarce  near  Tangier,  but 
common  about  Dar  el  Baidar  " ;  and  says  ".  they  cross  the  Straits 
in  spring,  returning  in  August  and  July." 

This  beautifully  marked  bird  is  the  most  common  Sand-Grouse 
on  the  Spanish  side,  although  very  local,  being  abundant  about 
the  edges  of  the  marisma,  where  they  nest  late  in  May,  as  also 
near  Granada.  Some  may  be  migratory ;  but  I  have  seen  them 
in  January  near  Seville.  I  never  saw  any  in  the  neighbourhood 
of  Gibraltar. 

The  flight  of  both  species  of  Sand-Grouse  is  very  powerful ; 
and  sometimes  they  go  to  such  a  height  that,  although  you  can 
hear  their  croaking  hoarse  call,  they  are  almost  out  of  sight. 
Becoming  excessively  tame  and  familiar  when  kept  in  con- 
finement, in  a  wild  state  they  are  very  difficult  to  approach 
without  a  stalking-horse,  and  when  obtained  are,  in  my  opinion, 
of  no  use  to  eat.  They  are  very  difficult  birds  to  skin,  the 
feathers  coming  out  like  those  of  a  Pigeon. 

Eggs  of  this  Pin-tailed  Sand-Grouse,  taken  near  Seville,  are  of 
a  reddish-buff"  colour,  marked  all  over  with  spots  of  reddish 
brown  and  light  grey. 

The  European  and  African  form  has  been  separated  as  a 
distinct  species,  P.  pyrenaicus,  because  they  are  of  richer  colour 
.than  Asiatic  birds  (!). 

Lower  breast  and  belly  white.     Chest  chestnut,  with  black  line  above  and 
below.     Two  centre  tail-feathers  much  elongated  and  pointed. 
Male.  Throat  black. 
Female.  Throat  white.     Length  13^  inches,  tail  5£  inches. 


THASIANID^.  237 

262.  Pterocles  arenarius  (Pallas).    The  Black-bellied  Sand- 
Grouse. 

Moorish.  El  Koudri.     Spanish.  Corteza. 

Favier  merely  mentions  that  this  Sand-Grouse  "  occurs  in  Dar 
el  Baidar." 

On  the  European  side,  this  large  species  of  Sand-Grouse, 
though  extremely  local,  is  resident  in  the  marisma  and  near 
Utrera,  nesting  late  in  May,  but  does  not  appear  near  Gibraltar. 
I  do  not  think  there  is  any  migration  of  this  bird.  The  different 
species  of  Sand-Grouse  lay  three  eggs  only,  of  an  elliptical  form, 
placed  on  the  bare  ground  without  any  nest ;  eggs  of  this  species 
taken  near  Seville  are  of  a  pale  cream-colour,  marked  all  over 
with  faint  spots  of  very  light  brown. 

Belly  uniform  black  ;  centre  tail-feathers  not  elongated. 

Male.  Throat  chestnut. 

Female.  Throat  yellowish  white.     Length  13  inches. 

Future  visitors  to  Morocco  should  look  out  for  Pterocles  seneyallus 
(Linnseus),  which  most  probably  occurs  in  the  southern  part  of  the  country, 
and  has  no  pectoral  zone.  Chest  and  upper  parts  covered  with  round 
black  spots.  Length  13  inches. 


Order   GALLINJE.       Family  PHASIANIDJE. 

263.  Caccabis  rufa  (Linnseus).     The  Western  Red-legged  or 
French  Partridge. 

Spanish.  Perdiz. 

This  Partridge  is,  throughout  Andalucia,  plentiful  and  resident, 
frequenting  the  monte  or  scrub,  not,  as  in  some  parts  of  England, 
being  found  in  cultivated  places.  Never  known  to  occur  on  the 
African  side  of  the  Straits,  it  is  not  even  found  on  the  Rock  of 
Gibraltar,  which  would  seem  rather  strange,  as  it  is  to  be  seen 
on  the  Queen  of  Spain's  Chair,  and  occasionally  on  the  plain 


238  PHASIANID.E. 

below  within  a  couple  of  miles  of  the  neutral  ground.  Mr.  Stark, 
in  February  1876,  found  "  large  packs  of  this  Partridge  on  the 
Sierra  Nevada  at  an  elevation  of  some  7000  feet." 

Almost  every  Spanish  sportsman,  or  cazador,  keeps  one  or  more 
of  these  birds  as  call-birds  (reclames},  each  wretched  Partridge 
being  confined  in  a  cage  which  is  so  small  that  the  unfortunate 
bird  has  scarcely  room  to  turn  round.  To  add  to  this  cruelty,  at 
certain  seasons  they  are  never  given  water,  as  it  is  supposed  to 
be  fatal  to  them ;  but  in  a  wild  state  they  drink  a  great  deal,  and 
during  the  scorching  month  of  August,  and  the  first  half  of 
September,  one  of  the  favourite  Spanish  methods  of  shooting 
them  is  to  make  a  hiding-place  (puesto)  near  their  drinking- 
haunt,  placing  call-birds  on  each  side  of  the  water  out  of  the  line 
of  fire —  so  that,  when  a  covey  comes  to  drink,  as  many  as  possible 
may  be  mowed  down  at  once  by  the  concealed  "  sportsman,"  who, 
throughout  the  whole  year,  regardless  of  the  season,  shoots  them 
whenever  he  can,  the  acme  of  his  diversion  being  to  shoot  a 
Partridge  from  the  nest.  However,  it  may  as  well  be  mentioned 
that  these  men  shoot  for  profit,  not  for  sport.  In  spite  of  this 
ceaseless  persecution,  Partridges  do  not  decrease,  which  is  truly 
wonderful.  Since  the  above  was  written  a  close-season  has  been 
established,  but  not  observed. 

Light-coloured  and  white  varieties  of  the  Eed-legged  Partridge 
are  of  not  unfrequent  occurrence :  for  some  years  in  succession 
there  was  a  white  covey  near  the  Guadiarro  on  the  road  to 
Gaucin.  They  breed  in  May,  and  the  eggs  vary  greatly  in  size, 
being  subject  to  the  same  variations  as  those  of  C.  petrota,  but 
usually  larger  than  the  eggs  of  that  bird,  although  sometimes 
smaller. 

Sexes  alike,  except  that  male  has  knobs  or  blunt  spurs  on  legs.  Black 
gorget  or  collar  from  bill  through  eyes,  enclosing  a  white  throat ;  dirty 
white  below  this  collar,  spotted  and  streaked  with  black  ;  bill,  eyelids,  and 
legs  red  ;  tail  chestnut,  of  fourteen  feathers.  Length  14  inches. 


CACCABIS  PETROSA.  239 

264.  Caccabis  petrosa  (J.  F.  Gmelin).  The  Barbary  Partridge. 

Moorish.  Hejel. 

"The  Barbary  Partridge,"  says  Eavier,  "is  resident  around 
Tangier,  and  very  common,  sometimes  perching  on  trees." 

This  species  is  far  more  common  in  proportion  in  Morocco 
than  C.  rufa  is  in  Spain,  and  chiefly  frequents  palmetto  scrub ; 
in  some  localities  it  is  so  numerous  that  it  would  be  quite  easy 
for  one  gun  to  bag  fifty  brace  in  the  day.  The  flesh  of  this 
Partridge  is  not  so  good  even  as  that  of  the  Common  Eed-leg, 
which  does  not  say  much  in  its  favour,  but  they  submit  to  capti- 
ivty  very  well,  and  may  be  kept  alive  in  coops  like  fowls,  to  be 
used  as  required — and,  after  being  fed  on  corn  for  a  month  or  so, 
improve  greatly  in  a  culinary  point  of  view. 

As  is  well  known  to  all  ornithologists,  the  Barbary  Partridge  is 
the  only  species  found  on  the  Rock  of  Gibraltar,  being  in  great 
numbers  there.  Sometimes  they  may  be  seen  sitting  on  the 
stones  within  a  few  feet  of  the  sea,  and  a  pair  or  two  used  even 
to  frequent  the  rocks  below  the  "rope  ladder"  at  Europa. 
Although  protected  from  guns  and  carefully  preserved  from  the 
attacks  of  human  beings,  they  suffer  considerably  from  the  number 
of  cats  which  abound,  and  are  also  preyed  on  by  Genets  and 
Eagles,  whilst  lizards  and  snakes  destroy  the  eggs  and  young. 

This  bird,  like  ether  Partridges,  is  very  noisy  at  dusk  :  in  the 
nesting-season  they  have  a  peculiar  long-drawn  croaking  cry, 
which  puzzled  me  for  a  long  time  before  I  could  make  out  from 
what  bird  it  came;  but  whether  the  male  only  thus  calls  I  do  not 
know,  though  I  suspect  such  to  be  the  case. 

They  commence  to  lay  about  the  15th  of  April.  The  eggs  are 
very  similar  to  those  of  C.  rufa,  and  vary  much  in  the  markings, 
some  being  quite  free  from  the  usual  small  freckles. 

The  Rev.  John  White  mentions  this  Partridge  as  being  plentiful 
at  Gibraltar  about  1770,  and  not  being  found  on  the  mainland  of 
Spain. 

This  species  is  at  once  distinguished  by  the  chestnut  collar  round  the 


240  PHASIANID.E. 

neck,  studded  with  small  white  spots,  and  is  also  a  smaller  bird  than  the 
Common  Red-leg,  besides  having  a  metallic-blue  tinge  on  the  wing-coverts. 
The  legs  are  not  always  red,  sometimes  being  a  pale  buff  colour.  Length 
inches. 


265.  Francolinus   bioalcaratus   (Linnaeus).     Double-spurred 

Francolin. 

This  is  no  doubt  the  species  mentioned  by  Mr.  Drake,  and 
occurs  as  far  north  in  Morocco  as  Rabat,  where  I  am  informed 
the  local  name  is  "  Ragh  "  ;  and  of  late  years  many  have  been  sent 
alive  from  Mogador  to  England. 

The  male  has  two  pairs  of  spurs.     The  sexes  are  alike  in  plumage. 

Above  brown  vermiculated  with  black  ;  chin  and  throat  white.  General 
colour  below  buff,  each  feather  with  a  black  racquet-shaped  shaft-streak. 
Length  12^  inches. 

266.  Coturnix  communis,  Bonnaterre.    The  Common  Quail. 
Moorish.  Summin.     Spanish.  Codorniz. 

.  Favier  states  that  the  Quail  is  very  abundant  on  passage  on 
the  Moorish  side  of  the  Straits,  many  remaining  to  breed,  the 
majority  crossing  over  to  Europe  during  March  and  April, 
returning  in  October  and  November. 

On  the  Spanish  side  of  the  Straits  it  appears  to  me  that  the 
chief  vernal  migration  of  the  Common  Quail  is  during  the  months 
of  March  and  April,  whilst  the  autumnal  passage  is  almost  entirely 
executed  during  the  latter  half  of  September,  at  that  time  their 
numbers  being  sometimes  almost  incredible. 

The  Andalucian  cazadores  profess  to  recognize  two  kinds  of 
Quail — those  which  are  migratory  and  called  "  Criollas,"  and 
those  which  are  resident  and  so  named  "  Castellanas"  There  is 
certainly  much  difference  in  the  colour  of  the  plumage  and  of  the 
legs,  the  Criollas  being  lighter-coloured  and  slightly  smaller  birds 
than  the  Castellanas,  which  are  very  dark ;  otherwise,  in  habits, 
note,  and  eggs,  there  is  no  difference,  although  at  a  glance  the 
resident  and  migratory  races  can  be  easily  distinguished. 


TURNICIDJE.  241 

There  are  a  great  number  of  these  resident  Quails,  which, 
throughout  the  winter,  seem  to  collect  together  and  haunt  certain 
favourite  spots,  these  places  never  being  without  them.  You 
may  kill  three  or  four  and  hunt  about  unable  to  find  more  ;  but 
go  to  the  same  place  in  a  few  days'  time,  and  you  will  find  that 
some  fresh  ones  have  taken  possession  of  the  ground. 

In  summer  Quail  are  universally  distributed  all  over  the  culti- 
vated country ;  in  autumn  the  best  place  to  shoot  them  is  in  the 
maize-fields  or,  rather,  stubbles.  Vast  quantities  are  caught  in 
the  spring  with  small  nets  by  the  aid  of  the  "  Quail-call "  (pitillo). 
The  birds  begin  to  call  their  love-note  about  the  9th  of  March ; 
after  that  time  their  "  quit  que-twit "  is  to  be  heard  on  every  side 
as  long  as  the  nesting-season  lasts.  They  commence  to  lay  in 
May ;  and  I  have  known  of  a  nest  with  eleven  eggs  taken  as  late 
as  the  17th  October. 

General  colour  buff,  marked  much  as  Partridge  above. 

Male.  Chin  and  throat  blackish. 

Female.  Chin  and  throat  buff  ;  larger  than  male.     Length  7  inches. 


Order  HEMIPODII.        Family  TURNICIDJB. 

267.  Turnix  sylvatica  (Desfontaines) .    The  Andalucian  Bush- 
Quail  or  Three-toed  Quail. 

Moorish.  Zerquil  (Favier).     Spanish.  Torillo. 

"  This  little  Quail  is  both  resident  and  migratory  in  the  vicinity 
of  Tangier,  and  is  a  much  less  common  bird  than  the  ordinary 
Quail;  those  which  migrate  pass  northwards  during  May  and 
June,  and  are  seen  on  the  return  passage  in  September  and 
October.  They  nest  in  July,  depositing  four  eggs  in  any  slight 
depression  of  the  ground,  often  among  corn.  The  young,  from 
the  moment  of  exclusion,  are  attended  by  both  male  and  female — 
all  remaining  together  in  parties  for  some  time,  in  the  same 
manner  as  Quails.  I  kept  a  female  bird  in  captivity  for  about 
thirteen  months,  feeding  it  on  millet  and  water.  This  bird  was 

E 


242  TURN1CIDJE. 

very  fond  of  eating  flies,  and  also  used  to  devour  the  ants  which 
came  into  its  cage  to  carry  away  the  dead.  Very  gentle  in  its 
character,  the  call  of  this  bird  was  very  triste :  it  cooed  day  and 
night,  much  in  the  manner  of  a  Turtle-Dove,  only  the  note  was 
more  subdued  and  lengthened.  I  have  reason  to  believe  that 
these  Quails  would  breed  in  captivity,  although  this  individual 
bird  did  not  lay." — Favier. 

On  the  Spanish  side  I  was  unable  to  detect  any  migration  of 
this  bird,  though  it  is  said  by  Andalucian  bird-catchers  and 
cazadores  to  be  migratory.  The  probability  is  that  they  are  so, 
but  yet  am  inclined  to  think  the  reverse,  as  they  are  found  in  the 
same  localities  in  equal  numbers  at  all  seasons  of  the  year. 

Near  Gibraltar  it  is  very  local  and  nowhere  plentiful,  apparently 
less  so  than  is  really  the  case,  for  they  are  difficult  birds  to  flush, 
and  if  put  up  once  will  rarely  rise  a  second  time.  Scattered  here 
and  there,they  chiefly  frequent  palmetto  (Chamcerops  humilis]  scrub, 
and  appear  to  be  most  common  near  the  coast,  being  more  abun- 
dant to  the  east  of  the  Queen  of  Spain's  Chair,  especially  about 
the  Lomo  del  Rey  and  a  place  called  Las  Agusaderas.  In  their 
flight  and  habits,  from  what  I  could  observe  of  them,  they 
resemble  the  Indian  Button-Quail  (Turnix  dussumieri). 

I  have  often  seen  them  among  the  rough  grass  and  bents  close 
to  the  sea-shore,  but  always  near  palmetto,  and  one  bird  in  parti- 
cular for  a  long  time  frequented  a  patch  of  thick  herbage  near 
the  mouth  of  the  "First  Iliver";  whenever  I  rode  by,  my  dog 
used  to  flush  it,  till  at  last  one  day,  wanting  a  specimen,  I  went 
purposely  to  shoot  the  bird ;  but,  of  course,  upon  this  occasion 
my  friend  was  not  to  be  found,  nor  did  I  again  see  one  there  for 
some  months. 

They  are  scarce  between  Algeciraz  and  Tarifa,  but  occur 
towards  Vejer,  and  are  tolerably  plentiful  on  the  palmetto- 
covered  high  ground  above  Casas  Viejas,  called  La  Mesa  ;  further 
than  this  I  did  not  meet  with  them  personally,  nor  could  I  obtain 
any  near  Seville. 


TURX1X  SYLVAT1CA.  243 

The  nest  is,  from  the  skulking  habits  of  the  birds,  extremely 
difficult  to  obtain.  I  never  had  the  good  fortune  to  find  one, 
but  had  one  lot  of  eggs  brought  to  me  from  near  San  Roque, 
on  the  6th  of  July,  1869.  The  finder  said  the  nest  was  under 
shelter  of  a  palmetto  bush,  and  merely  consisted  of  a  few  bits  of 
dried  grass.  These  eggs,  four  in  number,  which  is,  without 
doubt,  the  regular  complement  laid  by  all  the  Three-toed 
Quails  *,  were  very  slightly  incubated,  and  in  appearance  much 
resembled  those  of  the  Common  Pratincole,  Glareola  torquata, 
only  being,  of  course,  much  more  diminutive.  Later  in  July 
I  received  several  eggs  from  Mogador,  which  exactly  resembled 
the  Spanish  ones  ;  but  not  having  been  blown  and  being  hard 
sat-on,  the  shells  were  so  tender  and  rotten  as  to  be  useless. 
My  friend  Captain  S.  Reid,  of  the  Royal  Engineers,  informed  me 
that  he  had  the  luck  to  find  a  nest,  placed  in  grass  near  the 
shore  on  the  eastern  beach,  on  the  19th  of  May,  1873 ;  this  nest 
contained  four  incubated  eggs,  as  did  another  from  near  Tangier 
obtained  by  Olcese.  Verner  tells  me  that  on  the  llth  of  August 
two  eggs  hard  sat-on  were  brought  him  by  a  Spaniard,  who  had 
caught  the  old  bird  on  the  nest.  There  were  four  eggs,  but  he 
broke  two  whilst  catching  the  old  one. 

The  males  of  this  species,  and,  I  believe,  of  all  the  genus,  are 
very  much  smaller  than  the  females  ;  this  difference  is  so  striking 
that  the  cazadores  always  declare  there  are  two  species.  I  have 
at  different  times  kept  these  little  birds  alive,  and  sent  them  to 
England,  and  they  are  easily  reconciled  to  captivity,  becoming 
very  tame  and  confiding  pets ;  at  times  they  coo  in  a  moaning 
way,  whence  their  trivial  Spanish  name  of  torillo  or  little  bull. 
They  also  have  another  single  note,  much  like  that  of  the  female 
Quail  but  less  loud. 

This  bird  was  called  by  Latham  in  1783  the  Gibraltar  Quail. 

The  Anglicized  name  of  Hemipode  is  fallacious,  as  Nature 

*  Captain  Loche,^(?e  Canon  Tristram,  seems  to  have  been  singularly  fortunate 
in  finding  a  nest  with  seven  eggs  ('  Ibis,'  1859,  p.  80). 

E2 


244  KALL1D.E. 

produces  no  half-footed  animal,  whilst  the  absence  of  the  hind 
toe  or  hallux  is  not  peculiar  to  the  Turnicidse.  Bush-Quail, 
Three-toed  Quail,  or  Button-Quail,  as  in  India,  would  be  more 
appropriate  as  English  names. 

Sexes  alike.     Centre  of  chest  bright  rusty.     Length  :   male  6  inches ; 
female  7^  inches. 


Order  FITLIC AKI JE,          Family  RALLIDJE. 

268.  Crex    pratensis,    Bechstein.     The    Landrail    or   Corn- 

Crake. 

Moorish.  Zelga  (Favier).  Spanish.  Eey  de  los  Codornices 
(King  of  the  Quails),  Guia  de  los  Codornices  (Guide  of  the 
Quails),  Polluela  rubia. 

"  This  Crake  is  found  in  Morocco  on  passage,  crossing  the 
Straits  during  the  month  of  February,  returning  in  August, 
September,  and  October,  being  occasionally  obtained  throughout 
the  winter  months." — Favier. 

The  Landrail  does  not  seem  to  remain  in  Andalucia  during  the 
breeding-season,  as  I  never  heard  its  well-known  cry  ;  but  I  have 
seen  them  as  late  as  the  2nd  of  May.  They  are  not  obtained  in 
any  abundance,  but,  like  other  Rails,  are,  no  doubt,  more 
common  than  they  appear  to  be,  and  occur  most  frequently  in 
October  and  February,  being  sometimes  found  during  the  winter 
months. 

General  colour  above  brownish  buff,  each  feather  on  back  with  a  black 
centre  ;  axillaries  chestnut  ;  bill  shorter  than  head — as  in  the  three  following 
species.  Length  10^  inches. 

269.  Porzana  maruetta  (Leach).    The  Spotted  Crake. 
Spanish.  Polluela  (under  which  name  the  next  four  species  are 

also  included). 

"  This  bird  is  met  with  near  Tangier  during  passage,  but  not 
in  any  great  number,  and  is  the  most  common  of  the  family, 


PORZANA  PARVA.  245 

haunting  thick  beds  of  rushes  in  swamps  and  on  the  edges  of 
lakes  and  rivers." — Favier. 

The  Spotted  Crake  is  extremely  abundant  on  the  Spanish  side, 
being  more  numerous  than  the  Water-Rail,  and  is  most  frequent 
in  spring  and  autumn.  Many  remain  during  the  winter  months  ; 
and  they  are  found  also  in  the  breeding-season. 

General  colour  above  olive-brown  ;  each  feather  on  back  with  blackish 
centre,  and  spotted  with  white  ;  axillaries  barred  with  white ;  breast 
spotted.  Length  8|  inches. 

270.  Porzana  bailloni  (Vieillot).    Baillon's  Crake. 

Favier  says  of  this  bird: — "Very  rare;  I  only  met  with  one, 
in  1857."  But  I  obtained  two  near  Tangier  in  spring  of  1877, 
and  probably  it  is  common. 

Seldom  seen,  owing  to  its  skulking  propensities.  We  found 
this  prettily  marked  Crake  very  common  when  snipe-shooting 
at  Casas  Viejas  from  October  to  February.  We  also  obtained  it  at 
the  Laguna  de  la  Janda  in  May.  Many  are  resident,  breeding 
at  the  end  of  April,  when  they  make  a  small  nest  of  sedges  and 
grass  placed  at  the  edges  of  swamps,  laying  from  five  to  seven 
olive-brown  eggs  spotted  with  darker  brown,  very  like  Jay's  eggs. 

Above  dark  reddish  brown,  marked  with  black  ;  cheeks,  throat,  and 
breast  slaty  blue  ;  flanks  black,  barred  with  white  ;  more  white  streaks  and 
spots  on  back  than  in  next  species  ;  under  wing-coverts  mottled  with 
whitish. 

Young.  Chin  and  throat  white  ;  breast  and  belly  buff.     Length  7  inches. 

271.  Porzana  parva  (Scopoli).    The  Little  Crake. 

The  Little  Crake  is  not  noticed  by  Favier  as  occurring  in 
Morocco ;  and  on  the  Spanish  side  of  the  Straits  I  never  could 
succeed  in  meeting  with  any,  though  have  seen  specimens  said 
to  be  Andalucian,  and  Arevalo  records  them  from  province  of 
Malaga.  Owing  to  the  powers  of  concealment  which  these  small 
Rails  possess,  it  is  very  difficult  to  obtain  them,  and  impossible 
to  learn  much  of  their  habits.  Without  the  aid  of  a  good  dog  it 


246  RALL1D.E. 

is  very  hard  to  compel  them  to  rise ;  and  in  consequence  they 
appear  to  be  much  more  rare  than  they  really  are.  I  recollect 
finding  a  nest  of  this  or  Baillon's  Crake  situated  in  a  very  small 
isolated  patch  of  swamp  ;  and  instead  of  trying  to  snare  the  birds, 
I  stupidly  endeavoured  to  flush  them  with  three  good  water- 
dogs  ;  but  it  was  quite  useless ;  we  could  find  no  signs  of  them 
whatever  ;  so  the  unidentified  eggs  were  valueless,  as  the 
resemblance  is  so  great  between  the  eggs  of  this  and  Baillon's 
Crake,  that  unless  the  bird  be  obtained  it  is  impossible  to  tell  to 
which  species  they  belong. 

Much  like  last,  but  larger,  and  with  flanks  slate-grey  ;  white  spots  only 
on  centre  of  back  ;  under  wing-coverts  uniform.  Length  8  inches. 

272.  Rallus  aquaticus,  Linnseus.    The  Water-Rail. 

Spanish.  Rascon,  Rascon  de  agua. 

"  This  bird  is  found  on  passage  near  Tangier  in  about  the 
same  numbers  as  the  Landrail,  frequenting  the  edges  of  rivers 
and  swamps,  where  they  hide  up  in  the  sedges." — Favier. 

The  Water-Rail  is  very  common  in  all  suitable  localities  on  the 
Spanish  side;  and  their  croaking  frog-like  call  is  always  to 
be  heard  in  the  swampy  jungle  at  Casas  Viejas.  Being  to  a 
great  extent  a  migratory  bird,  they  are  most  common  in  winter ; 
but,  owing  to  the  cover  being  more  thin,  at  that  season  all  the 
Rails  and  Crakes  are  easier  to  obtain.  They  build  in  rushes  or 
sedges,  laying  about  the  20th  of  April.  On  the  13th  of  May  we 
found  two  nests,  from  each  of  which  Mr.  Stark  succeeded  in 
snaring  one  of  the  old  birds ;  these  nests,  built  entirely  of 
dry  sedge  and  lined  with  a  few  bits  of  dry  grass,  were  just 
raised  above  the  water,  and  measured  6  inches  in  height, 
depth,  and  diameter;  the  hollow  of  the  nest  was  4J  inches 
across  by  2J  deep.  Each  nest  contained  seven  eggs  hard 
sat-on — one  lot  being  of  the  usual  type,  the  other  resembling 
more  those  of  the  Spotted  Crake,  or,  rather,  looking  like 
miniature  Waterhen's  eggs  with  larger  blotches  than  usual. 


PORPHYRIO  (LERULEUS.  247 

Bill  reddish  at  base,  longer  than  the  head,  and  slender. ' 

Above   brown,  blotched  with  black  ;  below  slaty  blue  ;  axillaries,  belly, 

and  flanks  black,  barred  with  white  ;  legs  and  feet  brown.     Length  10 

inches. 

273.  Gallinula    chloropus    (Linnaeus).     The    "Waterhen    or 

Moorhen. 

Moorish.  Zelga-kahla  (Favier).     Spanish.  Polla  de  agua. 

The  Moorhen,  according  to  Favier,  is  "  resident  in  the  vicinity 
of  Tangier,  being  met  with  in  abundance ;  many,  however,  are 
migratory." 

It  is  needless  to  say  much  about  a  bird  so  well  known  as  our 
Common  English  Waterhen.  They  are  not  so  common  in 
Andalucia  as  the  Spotted  Crake ;  but  I  was  unable  to  detect  any 
migratory  habits  on  the  Spanish  side  of  the  Straits,  where  they 
are  tolerably  plentiful  and  generally  distributed  in  all  suitable 
localities,  often  being  seen  about  the  gardens  at  the  edge  of  the 
small  stream  at  Algeciraz  and  at  Vejer,  seeming,  as  in  England, 
to  be  fond  of  living  in  the  vicinity  of  houses  and  cultivation. 
They  nest  about  the  end  of  April. 

General  colour  above  olive-brown,  below  greyish  black  ;  scarlet  frontal 
plate  in  adult,  greenish  brown  in  the  young  ;  legs  and  feet  greenish. 
Length  13  inches. 

274.  Porphyrio  cseruleus  (Vandelli).    The  Purple  Waterhen. 

Moorish.  Kazeid  (Reid).     Spanish.  Mancon  aziil,  Calamon. 

"This  bird  is  chiefly  migratory,  and  not  common  near  Tangier, 
passing  north  during  the  months  of  February  and  March,  and 
returning  in  December  and  October.  They  are  occasionally  to 
be  seen  during  the  month  of  January,  but  not  every  year.  Those 
which  remain  for  the  breeding-season  construct  their  nests  in  the 
midst  of  wet  sedges  or  rushes,  depositing,  in  April,  from  three  to 
five  eggs.  When  these  birds  are  moulting  they  are  very  easy  to 
obtain,  as  they  lose  all  their  quill-feathers  at  once,  and  so  cannot 
fly." — Favier. 


248  RALLID^E. 

The  Purple  Waterhen  (a  large  and  very  handsome  bird)  is,  on 
the  Spanish  side  of  the  Straits,  very  irregular  in  its  appearance 
both  as  to  time  and  locality.  In  some  years,  during  January  and 
February,  they  are  to  be  seen  near  Gibraltar  in  situations  where 
they  do  not  occur  at  any  other  time,  and  are  then,  doubtless,  on 
migration. 

In  wet  seasons  they  nest  near  Casas  Viejas  in  April,  in  the 
Soto  Malabrigo,  where  I  have  shot  them  as  late  as  the  27th  of 
October.  They  are  very  difficult  to  flush  without  a  dog ;  when 
they  rise  they  make  a  flapping  noise,  and  with  a  heavy  flight 
merely  take  refuge  in  the  nearest  thick  patch  of  rushes  or  wet 
sedgy  jungle,  whence,  from  being  Crake-like  in  their  habits,  it  is 
almost  impossible  to  make  them  rise  a  second  time.  Some  are 
to  be  found  in  a  few  places  at  the  edge  of  the  marisma  of  the 
Guadalquivir.  The  nest  resembles  that  of  the  Common  Coot ; 
and  the  eggs,  which  are  richly  coloured,  are  laid  towards  the  end 
of  April. 

The  gizzards  of  those  which  I  have  examined  contained  nothing 
but  vegetable  matter  (grass,  seeds  of  rushes,  &c.),  with  a  good 
deal  of  coarse  gravel. 

Purplish  blue,  darkest  on  tail  ;  tail-coverts  white.  Bill  and  frontal 
shield  bright  red  ;  legs  and  feet  red.  Length  16-18  inches. 

275.  Fulica  atra,  Linnaeus.    The  Common  Coot. 

Moorish.  El  Ghor  (Favier).     Spanish.  Mancon,  Gallareta. 

"  This  Coot  is  resident  near  Tangier,  but  is  not  very  numerous, 
often  consorting  with  Fulica  cristata.  Some  are  migratory, 
passing  northwards  during  the  months  of  January  and  February, 
and  returning  in  August  and  September." — Favier. 

We  found  the  Common  Coot  abundant  near  Tetuan  in  March  ; 
and  they  are  common  on  the  Spanish  side,  particularly  in  winter, 
when  very  large  flocks  appear,  and  some  are  then  to  be  seen  on 
the  inundation  at  the  North  Front.  They  breed  about  the 
middle  of  April  in  all  large  swamps,  particularly  at  the  Laguna 


GRUID.E.  249 

de  la  Janda,  wrhere,  though  the  nests  are  numerous,  it  is  almost 
impossible  to  see  the  birds,  owing  to  the  density  of  the  grass  and 
rushes. 

Front  toes  with  rounded  lobes. 

General  colour  slaty  black  ;  secondaries  tipped  with  white,  forming  a 
wing-bar  ;  broad  white  shield  on  forehead  ;  legs  bluish  grey  ;  iris  crimson. 
Length  16  inches. 

276.  Fulica  cristata,  Gmelin.    The  Red-lobed  Coot. 

Spanish.  Gallareta. 

"  This  bird  is  both  resident  and  migratory  near  Tangier. 
Those  which  migrate  return  from  the  north  in  September. 
The  nest  and  eggs  resemble  those  of  Fulica  atra,  with  which 
species  they  associate,  but  are  much  more  numerous." — Favier. 

This  Coot  breeds  at  Ras  el  Doura  in  numbers  about  the  20th 
of  April;  and,  as  above  mentioned,  the  eggs  are  not  to  be 
distinguished  from  those  of  the  Common  Coot  (F.  atra) ;  so 
unless  the  bird  be  snared  on  the  nest,  the  eggs  cannot  be  said  to 
be  identified. 

In  Andalucia  they  have  been  obtained  about  the  Laguna  de  la 
Janda,  but  are  much  more  common  further  east. 

I  prefer  to  call  this  bird  Red-lobed,  as  it  certainly  is  not  crested,  and 
resembles  the  Common  Coot  in  all  respects  except  having  two  red  lobes  or 
caruncles  on  the  white  frontal  shield  and  not  having  a  white  wing-bar  or 
white  tips  to  the  secondaries.  Length  16  inches. 


Order  ALECTORIDES.         Family  aRUID£). 

277.  Grus  communis,  Bechstein.    The  Common  Crane. 

Moorish.  Gharnook  (Favier).     Spanish.  Griilla. 

"  This  Crane,  common  in  flocks,  is  found  in  Morocco  only 
during  the  winter  season,  arriving  in  October  and  November; 
they  leave  for  the  north  in  February." — Favier. 

On  the  Moorish  side  of  the  Straits  the   Common  Crane  does 


250  GRUID.E. 

not  appear  to  remain  to  nest,  as  I  looked  in  vain  for  them  in  the 
marshes  there  during  the  month  of  April. 

On  the  Spanish  side  some  thirty  to  forty  pairs  breed  in  the 
district  (comprising  many  thousand  acres)  which  extends  from 
Tapatanilla  along  the  Laguna  de  la  Janda  to  Vejer,  and  thence 
eastward  to  Casas  Viejas.  These  birds  commence  to  lay  about 
the  last  week  in  April  (but  Verner  records  a  nest  with  two  fresh 
eggs  on  the  4th  of  April,  1876),  constructing  their  nests  somewhat 
like  those  of  the  Swan,  of  sedges,  grass,  and  rushes.  The  nests 
vary  much  in  size,  some  being  quite  five  feet  across,  others 
perhaps  not  much  more  than  eighteen  inches :  some  are  deep, 
and  stand  high  up ;  others  are  almost  level  with  the  water,  in 
which  they  are  always  built.  The  nest  is  always  placed  among 
sedges  or  rushes  sufficiently  short  for  the  bird,  when  standing  up, 
to  be  able  to  see  around,  and  is  never  built  in  tall  reeds.  They 
are  very  easy  to  find,  as  the  old  birds  never  fly  direct  to  the  nest, 
but  alight  some  twenty  or  thirty  yards  away  and,  walking  up  to 
it,  form  regular  tracks  like  a  cattle-path  ;  so  by  following  one  of 
these  tracks  you  may  be  sure  of  finding  the  nest :  nor  do  the  old 
birds  fly  straight  away  from  the  nest,  but  walk  off  quietly  to  the 
end  of  one  of  these  paths  and  then  take  wing.  When  approached 
while  sitting  on  the  nest,  the  bird  slips  off,  crouches  down,  and 
runs  away  for  some  yards. 

Mr.  Stark  watched  a  pair  of  Cranes  for  two  or  three  days  from 
a  hill  which  directly  overlooked  a  marsh  where  the  process  of 
building  was  being  carried  on  ;  and  he  informed  me  that  only 
one  bird  worked  at  a  time,  the  other  standing  on  guard.  The 
nests  are  never  in  very  close  proximity  to  each  other,  and  never 
contain  more  than  two  eggs,  placed  side  by  side  so  as  almost  to 
touch,  both  the  small  ends  pointing  in  the  same  direction. 
Sometimes  the  second  egg  is  not  laid  until  two  or  three  days 
after  the  first,  and  they  differ  much  in  size  and  shape  in  different 
nests  ;  but  the  pair  in  a  nest  are  always  alike  in  size,  shape,  and 
colour,  which  varies  from  light  buff  to  an  olive-brown,  sometimes 


GRUS  COMMUNIS.  251 

marked  all  over  with  brown  and  reddish-brown  spots,  generally 
thickest  at  the  larger  end ;  but  some  eggs  are  almost  spotless. 

These  noble-looking  birds  are  very  much  harassed  during  the 
breeding-time  ;  and  being  said,  I  believe  correctly,  not  to  lay  a 
second  time  in  the  season  after  the  nest  has  been  robbed,  they 
will,  I  am  afraid,  soon  cease  to  breed  near  Casas  Viejas,  as  they 
have  almost  done  in  the  marisma  of  the  Guadalquivir,  owing  to 
ceaseless  persecution.  According  to  what  one  hears,  they  used 
years  ago  to  nest  there  in  great  numbers.  However,  it  is  the 
same  story  everywhere:  all  wild  birds  are  in  Europe  certainly 
decreasing  at  their  breeding-places,  owing  to  egging,  drainage, 
and  what  is  termed  civilization  ;  and  soon  it  will  come  to  nothing 
but  Dorking  Fowls  and  domestic  Pheasants. 

These  Andalucian-breeding  Cranes  are  very  largely  reinforced 
by  the  autumn  migration,  which  arrives  early  in  October ;  and 
they  then  form  immense  bands  of  from  two  to  three  hundred  in 
number,  though  generally  they  keep  in  smaller  lots  of  from  five 
to  thirty  or  forty.  Those  which  do  not  remain  to  nest,  pass 
north  in  March.  On  the  llth  of  that  month,  in  1874,  Mr.  Stark 
and  myself  had  the  pleasure  of  seeing  them  on  passage ;  and  a 
grand  and  extraordinary  sight  it  was,  as  flock  after  flock  passed 
over  at  a  height  of  about  two  hundred  yards — some  in  single  line, 
some  in  a  V-shape,  others  in  a  Y-formation,  all  from  time  to  time 
trumpeting  loudly.  We  watched  them  for  about  an  hour  as  they 
passed,  during  which  time  we  calculated  that  at  least  four 
thousand  must  have  flown  by.  This  was  early  in  the  morning, 
and  we  were  obliged  to  continue  our  journey ;  but  when  we  lost 
sight  of  the  vega  of  Casas  Viejas,  over  which  the  Cranes  were 
passing  in  a  due  northerly  direction,  there  appeared  to  be  no 
diminution  in  their  number,  and,  as  my  friend  remarked,  "  One 
would  not  have  believed  there  were  so  many  Cranes  in  all 
Europe."  These  birds  must  have  crossed  the  Straits  from  Africa 
that  morning,  the  place  over  which  we  saw  them  passing  being 
not  twenty  miles  in  a  direct  line  from  Tarifa,  and  a  line  drawn 


252  GRUID^E. 

in  the  direction  from  which  the  birds  came  would  have  fallen  a 
little  to  the  west  of  that  town.  At  Adra  on  the  16th  of  March, 
1882,  we  saw  eleven  Cranes  passing  north  at  a  great  height. 

In  the  winter  of  1870-71  1  found  that  Cranes  could  be  easily 
shot  in  the  evening  by  waiting  for  them  in  the  swamps  where  they 
resort  to  pass  the  night,  as  they  "  flight "  earlier  than  Ducks ; 
and  although  in  the  daytime  no  bird  is  so  wide  awake,  they  are 
quite  stupid  in  the  dusk,  flying,  if  you  keep  perfectly  still,  within 
a  few  yards.  It  is,  however,  a  barbarous  shame  to  shoot  such  a 
fine  and  noble  bird.  Although  the  Spaniards  gladly  take  them 
to  eat,  to  my  mind  their  flesh  is  coarse  and  worthless  ;  but  in 
India,  where  they  feed  much  on  grain  and  on  rice-stubbles,  they 
are,  on  the  contrary,  much  sought  after  for  the  table.  One  or 
two  shot  in  the  evening  near  Casas  Viejas  had  been  eating 
beetles  and  insects,  which  in  winter  seems  their  chief  food. 
They  do  a  great  deal  of  damage  to  beans  when  ripening  and  to 
newly  sown  grain  of  all  descriptions. 

General  colour  slate-grey  ;  inner  secondaries  bluish  black,  forming 
elongated  plumes. 

Adults  with  a  red  warty  patch  on  crown.  In  young  this  is  absent,  and 
the  grey  feathers  on  back  are  mixed  with  brown.  Length  48  inches. 

278.  Grus  virgo  (Linnseus).    The  Demoiselle  Crane. 

Spanish.  Grulla  moruna  (Seville). 

The  only  note  which  Favier  has  relative  to  this  handsome 
Crane  is  that  "it  is  scarce  and  seldom  obtained  near  Tangier, 
passing  northwards  without  making  any  stay,  during  March, 
April,  and  May." 

Favier's  successor  at  Tangier  evidently  considered  this  species 
a  rare  bird ;  for  he  asked  fifteen  dollars  (over  £3)  for  a  specimen, 
and  at  that  price  it  was  likely  to  continue  for  some  time  on  his 
hands.  He  stated  that  the  local  name  was  "  Bou-gernan  "  (father 
of  thistles) ;  but  if  the  bird  be  as  rare  as  Favier  implied,  how 
could  it  bear  a  local  name  ? 


OTIDIB.E.  253 

On  the  Spanish  side  I  failed  to  meet  with  this  Crane  near 
Casas  Viejas,  but  strongly  suspect  that  in  some  seasons  they  nest 
there ;  indeed  a  pair  of  Cranes'  eggs  that  were  brought  to  me 
were  so  small  that  I  could  not  refer  them  to  the  Common  Crane, 
but  could  of  course  obtain  no  reliable  information  about  them. 
Indeed,  an  egg  unidentified  is  worse  than  useless  to  the  ornitho- 
logist ;  and  unless  the  collector  takes  and  identifies  specimens 
himself,  he  had  better  not  keep  them. 

In  the  marisma  of  the  Guadalquivir  there  is  no  doubt  that 
in  former  years  the  present  species  used  frequently  to  breed. 
Specimens  are  often  to  be  obtained  near  Seville  during  March, 
April,  and  the  early  part  of  May,  and  again  in  August.  Judging 
from  this,  they  must  nest  somewhere  a  little  further  north. 

General  colour  bluish  grey,  sides  of  the  head,  throat,  neck,  the  lengthened 
pointed  feathers  on  upper  breast,  and  quill-feathers  black;  from  each  eye 
a  tuft  of  lengthened  white  feathers  extending  backwards.  Length  30 
inches. 

OTIDIDJE.     No  hind  toe. 

279.  Otis  tarda,  Linnaeus.    The  Great  Bustard. 

Spanish.  Abutarda. 

Favier  states  that  this  Bustard  "  occasionally  migrates  to 
Morocco  during  winter  from  the  European  side  of  the  Straits, 
but  very  rarely  remains  for  the  breeding-season." 

I  have  seen  one  or  two  specimens  obtained  near  Tangier,  and 
Mr.  Drake  also  mentions  one ;  whilst  Mr.  Meade-Waldo  kindly 
informs  me  that  in  the  spring  of  1892  he  met  with  many,  and 
saw  one  band  of  twenty-three.  He  found  a  male  on  a  small 
plain  which  when  fired  at  flew  round  and  round,  not  seeming  to 
like  to  pass  over  the  hills ;  at  last  he  got  right  up,  and  went 
away  like  a  goose. 

On  the  Spanish  side  of  the  Straits  the  Great  Bustard  is  first 
to  be  met  with  near  Gibraltar  on  the  plain  below  Facinas,  about 
ten  or  twelve  miles  from  Tarifa ;  northwards  from  there  they  are 


254  OTIDID.E. 

to  be  seen  in  gradually  increasing  numbers  all  along  the  vega  of 
the  Laguna  de  la  Janda  up  to  Casas  Viejas  and  along  by  Medina 
Sidonia  to  the  plains  which  lie  towards  Jerez  and  the  marisma 
of  the  Guadalquivir.  They  are  found  in  abundance  along  the 
line  of  railway  to  near  Utrera,  being  more  plentiful  in  the  open 
corn-growing  country  about  Marchena,  Coronil,  and  Carmona 
than  in  any  other  district  that  I  have  visited ;  thence  to  the 
north  side  of  the  Guadalquivir  they  are  also  common,  particularly 
about  Brenes  and  Alcala  del  Rio,  sometimes  appearing  very  near 
to  Seville — in  fact  in  all  open  country  the  Great  Bustard  may  be 
^expected  to  be  seen  in  varying  numbers.  I  never  saw  any  very 
large  flocks,  but  occasionally  have  observed  as  many  as  fifty 
together ;  from  ten  to  twenty-five  is  the  usual  number  seen  at 
once.  I  did  not  detect  any  migration  of  the  Great  Bustard, 
which  is  singular,  as  in  the  Crimea  some  very  large  flights 
appeared  on  passage  in  autumn;  however,  doubtless  they 
migrate,  as  they  are  occasionally  seen,  and  in  18£9  two  were  shot 
very  near  Gibraltar ;  certainly  they  shift  their  ground  in  Spain, 
as  they  are  most  numerous  about  the  vega  of  La  Janda  during 
the  breeding-season,  which  tends  to  the  belief  that  they  pass  over 
to  the  African  side  of  the  Straits.  In  a  wild  state  they  feed 
chiefly  on  grass  and  vegetable  substances  ;  and  when  kept  tame 
they  will  devour  any  amount  of  grasshoppers,  insects,  &c.,  the 
best  food  for  them,  however,  being  cabbage-leaves. 

Bustards  are  very  difficult  to  approach  except  by  some  such 
stratagem  as  driving  a  cart  near  them,  when  they  seem  to  fear 
no  danger ;  but  the  best  way  of  obtaining  them  is  by  driving 
when  the  corn  is  sufficiently  high  to  shelter  the  guns,  which  it 
usually  is  by  the  end  of  March.  It  is  necessary  to  have  for  a 
Bustard-drive,  with  any  chance  of  success,  at  least  four  guns ;  the 
more  the  better ;  and  as  the  birds  fly  almost  always  well  within 
shot  of  the  ground,  they  are  very  easily  killed  if  they  pass  over 
the  spot  where  a  gun  is  posted.  Indeed,  considering  the  size  of 
the  bird,  it  is  wonderful  how  light  a  wound  will  bring  one  down. 


OTIS  TARDA.  255 

I  have  seen  an  old  male  when  winged  and,  as  it  were,  brought  to 
bay,  turn  round  and  charge  his  pursuer.  This  diversion  of 
Bustard-driving  is  rather  an  expensive  pastime,  and  often,  like 
the  Irishman's  pig,  they  refuse  to  be  driven  in  the  required 
direction ;  so,  beyond  the  novelty  of  the  affair,  and  the  sight  of 
so  many  of  these  truly  noble  birds  on  the  wing,  there  is  nothing 
very  exciting  in  the  sport,  if  it  may  be  so  called ;  and,  as  in  all 
"  driving,"  there  is  no  sporting-skill  required  on  the  part  of  the 
shooter.  .  Nevertheless  the  thorough  enjoyment  of  the  bright  and 
glorious  climate,  and  the  sense  of  freedom  to  go  where  you  wish 
without  being  warned  off  as  a  trespasser,  and  last,  but  not  least, 
the  sociable  nature  of  the  "  entertainment,"  render  a  few  days' 
Bustard-driving  very  agreeable :  they  ought  not  to  be  shot  after 
the  third  week  in  April,  as  they  then  begin  to  lay;  but  most 
of  those  killed,  I  am  sorry  to  say,  are  destroyed  during  the 
breeding-season.  There  is,  morever,  the  lamentable  fact  that  the 
game  is  not  very  much  worth  having,  the  flesh  being  dry  and 
coarse ;  at  least  such  has  been  the  case  with  almost  all  that  I 
have  tasted. 

One  circumstance  in  favour  of  these  excursions  after  Bustard 
is  that  they  are  easily  managed  from  Seville  by  starting  by  the 
early  morning  train  and  returning  late  in  the  evening,  and  there 
is  no  trouble  in  having  to  search  for  uncomfortable  country 
quarters  for  the  night. 

The  man  we  always  employed  to  drive  was  one  Molino,  of 
Algaba,  a  small  village  or  pueblo  on  the  Guadalquivir,  above 
Seville,  a  wiry  active  little  fellow,  but  with  an  enormous  capacity 
for  meat  and  wine.  He  never  attempted  to  drive  the  birds  with 
more  than  three  men,  including  himself;  but  his  skilful  manage- 
ment, owing  to  his  knowledge  of  the  birds  and  the  ground,  and 
consequently  of  the  route  they  would  take,  was  something 
marvellous.  Molino  was  always  employed  by  some  Sevillanos, 
who  regularly  several  times  in  the  spring  go  out  dressed  in  green, 
like  Free  Foresters  at  the  Crvstal  Palace :  and  a  suitable  dress  it 


256  OTIDID.E. 

is ;  for  the  colour  being  that  ot  the  corn  or  grass  in  which  the 
tirador  lies  hid,  the  Bustards  are  not  nearly  so  likely  to  notice 
the  ambuscade.  The  day  these  verdant  gentlemen  choose  for 
their  "  funcion"  of  Bustards  is  invariably  Sunday  ;  and  sometimes 
they  succeed  in  killing  a  dozen  birds  in  the  day,  usually  about 
the  vicinity  of  Las  Alcantarillas.  In  August,  near  Casas  Viejas, 
and,  I  am  informed,  also  in  other  places,  the  Spaniards  ride  down 
Bustards  with  dogs,  continually  flushing  them  till  they  are 
exhausted ;  but  probably  young  birds  only  are  thus  caught. 
They  are  also  said  to  tire  out  the  Red-legged  Partridge  in  the 
same  way.  This  is  very  likely,  as  I  have  seen  these  Partridges 
when  driving  in  Norfolk,  after  being  flushed  two  or  three  times, 
allow  themselves  to  be  caught  quite  uninjured. 

The  Great  Bustard  is  easily  noticed  when  on  the  ground  where 
the  cover  is  not  too  high  to  hide  them ;  and  at  times  their  size 
appears  gigantic  as  they  fly  with  a  slow,  measured,  laborious- 
looking  flight ;  but  their  pace  is  much  faster  than  it  appears  to 
be ;  and  when  put  up  they  often  fly  a  distance  of  at  least  two 
miles.  They  have  great  power  of  concealment ;  and  I  remember 
an  instance  of  one  which  was  unable  to  fly,  from  some  injury  he 
had  received  in  one  of  his  wings.  We  saw  this  bird  in  a  corn- 
field of  some  forty  acres,  and  forming  line  we  tried  to  catch  him  ; 
but  he  suddenly  disappeared  in  the  corn,  which  was  not  more 
than  two  feet  high,  and  in  many  places  not  a  foot  high.  We 
spent  an  hour  in  vain  hunting  for  him  with  a  dog ;  so,  after 
beating  the  whole  field  over  more  than  once,  we  sat  down  in 
view  of  the  ground  to  eat  luncheon.  In  about  a  quarter  of  an 
hour  the  Bustard  appeared  some  three  hundred  yards  off  in  the 
middle  of  the  corn ;  so  I  went  straight  at  him,  running  as  hard 
as  I  could.  He  again  disappeared ;  but  going  on  I  suddenly 
spied  something  white  running,  as  it  were,  close  to  the  ground, 
and  rushing  after  it,  up  jumped  the  Bustard,  running  along  and 
flapping  his  wings ;  unable  to  catch  him  (for  he  ran  as  fast  as  I 
could),  I  was  compelled  to  shoot  him,  a  magnificent  Barbon  of 


OTIS  TETRAX.  257 

about  thirty  pounds  weight.  I  could  not  have  believed  so  large 
a  bird  could  crouch  so  low  and  at  the  same  time  make  such 
good  running. 

About  Casas  Viejas  a  few  Bustards  are  to  be  found  near  the 
banks  of  the  rivers  Barbate  and  Celemin,  where,  from  the  nature 
of  the  ground,  which  is  intersected  by  the  windings  and  branches 
of  these  rivers,  which  are  nearly  dry  in  August,  it  is  sometimes 
possible  to  stalk  them  under  cover  of  the  banks,  and  is  very 
good  ground  to  have  them  driven  over ;  but  the  Spaniards  then 
could  not  be  made  to  understand  such  work. 

The  Great  Bustard  nests  in  corn  or  grass  early  in  May,  laying 
two  olive-brown  eggs  marked  with  spots  and  blotches  of  dark 
brown.  Verner  says  he  has  twice  come  across  nests  containing 
four  eggs,  but  the  Spaniards  say  two  is  the  complement  laid  by 
each  hen,  and  if  more  than  that  number  are  found  in  a  nest  then 
two  hens  have  laid  together.  He  found  a  nest  as  late  as  the 
27th  of  June,  1879,  containing  three  fresh  eggs.  They  appear  to 
be  polygamous,  whilst  the  Little  Bustard  undoubtedly  pairs. 
The  gular  pouch,  which  always  exists  in  the  old  males  or 
Barbones,  is  sometimes  very  large  and  necked  in  the  middle, 
somewhat  like  an  hour-glass,  the  lower  part  being  the  largest ; 
this  shape  is  not  constant  and  perhaps  the  result  of  age. 

3 [ale.  Head  bluish  grey  ;  general  colour  above  reddish  buff,  with  black 
bands  on  each  feather  ;  wings  white  or  greyish,  except  the  dark  primaries  ; 
tuft  of  long  bristly  white  feathers  from  base  of  bill ;  band  of  rich  chestnut 
across  breast,  with  grey  band  underneath,  below  this  white. 

Female  and  young  male.  Without  bristles  on  throat  or  band  on  chest,  and 
back  more  banded  with  black.  Length:  male  45  inches,  female  36.  Greatest 
weight  of  a  male  37  Ib.  (Verner},  others  have  recorded  to  36  Ib. 

280.  Otis  tetrax,  Linnceus.    The  Little  Bustard. 

Moorish .  Bou-zerat  (Father  of  the  armourer :  Ead,  "  thim- 
derer  "),  Saf-saf ;  Sirt,  sirt  (from  call  or  note).  Andalucian.  Sison, 
Francolino. 

"  The  Little  Bustard  is  abundant  in  the  vicinity  of  Tangier  in 


258  OTID1D.E. 

small  flocks,  which  are  very  wild  and  wary.  They  migrate  to  the 
north  during  the  months  of  April  and  May,  returning  in  October 
and  November.  In  addition  to  these  migratory  birds,  great 
numbers  are  resident  during  the  nesting-season.  The  males  do 
not  attain  the  full  breeding-plumage  until  their  third  year,  and 
by  October  regain  the  dress  of  the  females." — Favier. 

My  experience  of  the  breeding-plumage  of  the  Little  Bustard 
is  rather  different  from  the  above ;  for,  as  far  as  I  have  been  able 
to  ascertain,  the  males  lose  the  black  markings  of  their  nuptial 
dress  by  the  end  of  August,  if  not  before.  I  could  not  make  out 
the  exact  period  ;  but  never  saw  a  black-marked  male  which  had 
been  killed  after  the  middle  of  August.  The  adult  males  never 
lose  the  minutely  marked  or  vermiculated  plumage  on  the  back, 
which  part  in  the  females  and  young  males  is  more  spotted  or 
blotched,  like  the  feathers  of  the  Great  Bustard.  I  found  the 
Little  Bustard  equally  common  in  Morocco  as  in  Andalucia  on 
all  open  low  cultivated  ground.  On  the  dead  level,  or  vega,"  of 
the  Barbate  near  Casas  Viejas  at  times,  in  early  autumn,  they 
positively  swarmed  in  flocks  sometimes  of  as  many  or  more  than 
a  hundred  together,  frequenting  this  flat  ground  till  swamped  by 
the  rains.  They  then  resorted  to  higher  and  undulating  ground, 
and  these  large  flocks  gradually  dispersed  and  broke  up  into  lots 
of  from  five  or  six  to  twenty  in  number.  They  are,  as  Favier 
remarks,  exceedingly  wary,  except  during  the  breeding-season 
and  in  the  month  of  August.  At  other  times  the  only  way  to 
obtain  them  was  by  driving,  very  uncertain  work,  as,  unlike  the 
Great  Bustard,  they  usually  rose  high  up  at  once,  and  their 
power  and  rapidity  of  flight  is  astonishing  for  their  size  and 
weight. 

They  were  often  to  be  seen  flying  somewhat  like  Golden  Plover, 
twirling  and  twisting  about  at  a  great  elevation ;  and  sometimes 
I  watched  them  rise  and  go  to  such  a  height  that  it  would  have 
been  difficult  to  tell  what  birds  they  were  unless  I  had  seen  them 
fly  up  from  the  ground. 


OTIS  TETBAX.  259 

During  August,  when  very  hot,  between  eleven  and  four,  they 
lie  "  like  stones  "  in  long  grass,  requiring  a  dog  to  flush  them ; 
but  the  heat  is  then  so  excessive  that  one  is  almost  as  likely  to 
get  a  sunstroke  as  a  Little  Bustard,  and  I  myself  could  never 
stand  such  work. 

The  nearest  place  to  Gibraltar  that  these  birds  are  seen  is  on 
the  plain  between  Los  Barrios  and  Palmones,  where  occasionally 
in  autumn  and  winter  a  few  appear  ;  but  they  are  too  much 
bullied  by  Gibraltar  sportsmen  to  remain  there  long.  The 
Moorish  names  given  above  are  significant  of  the  rattling  noise 
which  the  Little  Bustard  makes  in  rising ;  and  when  the  flock  is 
large  this  can  be  heard  a  very  long  way  off.  There  is  none  of 
this  sound  of  the  wings  in  the  rising  of  the  slow-flying  Great 
Bustard. 

When  on  the  wing,  the  Little  Bustard,  except  when  at  a  great 
height,  may  always  be  recognized  by  its  white  or,  rather,  pied 
appearance,  caused  by  the  greater  part  of  the  wings  being  white. 
"When  these  are  closed,  and  they  are  settled  on  the  ground,  this 
white  does  not  show,  and  they  are  very  difficult  to  make  out  or 
notice,  particularly  as  they  usually  frequent  ground  which  has 
some  cover  (in  the  shape  of  weeds,  thistles,  or  grass).  In  the 
breeding-season  they  keep  entirely  among  thick  herbage,  and  at 
that  time  I  never  could  get  a  sight  of  one  on  the  ground. 

In  my  opinion  there  is  no  bird  better  food  than  the  Little 
Bustard,  and  they  are  equal  to  the  Indian  Florican  in  this  respect, 
which  is  saying  a  great  deal  for  them. 

The  male  Little  Bustard  in  the  breeding-season  has  a  most 
peculiar  call,  which  can  be  easily  imitated  by  pouting  out  and 
pressing  the  lips  tight  together  and  then  blowing  through  them  ; 
the  birds  when  thus  calling  seem  to  be  close  to  you,  but  are  often 
in  reality  half  a  mile  off.  They  must  possess  powers  of  ventri- 
loquism, as  I  have  often  imagined  that,  they  were  quite  close  to 
me,  and  upon  hunting  the  spot  with  a  dog  found  no  signs  of 
them  anywhere  near ;  indeed,  at  that  season  it  is  sometimes  as 

s2 


260  .  OTIDID.E. 

difficult  to  make  them  rise  as  a  Landrail.  They  nest  in  the 
beginning  of  May,  laying  three  shiny  smooth  olive-green  eggs, 
more  or  less  blotched  with  dark  brown,  which  are  placed  among 
the  corn  or  long  grass. 

Male  in  spring  till  August.  General  colour  above  sandy  brown  ;  throat 
and  cheeks  bluish  black,  with  two  white  gorget-marks  across. 

Adult  male.  Back  vermiculated  with  black,  but  otherwise  in  winter 
resembles  the  female. 

Female.  Back  spotted  or  Uotclied  with  black  on  sandy-brown  ground- 
colour ;  underparts  white  ;  iris  golden  yellow.  Length  16-17  inches. 

281.  Eupodotis     arabs      (Linnaeus).       The     North  -  African 

Bustard. 

This  large  Bustard  was  obtained  by  Mr.  Tyrwhitt-Drake  in  the 
north  of  Morocco ;  and  towards  the  south,  about  Mogador,  it  is 
stated  to  be  common. 

Larger  than  the  Great  Bustard  (Otis  tarda),  it  has  the  entire 
back  covered  with  those  delicate  vermiculated  feathers,  sandy 
brown  crossed  with  fine  lines  of  deep  brown  almost  black,  which 
are  so  valuable  for  artificial  flies.  Similarly  marked  feathers  are 
found  in  the  other  species  of  Eupodotis  in  India,  the  Cape,  and 
Australia. 

282.  Houbara  undulata   (Jacq.).    The  Western  Houbara  or 

Ruffed  Bustard. 

This  Bustard  is  not  mentioned  by  Favier;  but  I  saw  one 
specimen  which  had  been  obtained  near  Tangier  in  August ; 
further  south  it  is  stated  to  be  frequently  met  with. 

In  Andulacia  is  a  rare  straggler  and  has  occurred  near  Malaga. 

Above  blotched,  not  vermiculated.  Crest  of  white  feathers  ;  black  and 
white  ruff  on  sides  of  neck.  Length  26-30  inches. 


GLAREOLID.E.  261 


Order  L  IMI  C  0  L  53  .         Family  GL AREOLID.E. 

283.  Glareola  pratincola  (Linnaeus).    The  Pratincole. 

Moorish.  Gharrak  (Favier),  Harrak-diad  (Drummond  -  Hay, 
P.  Z.  S.  1840).  Spanish.  Canastela. 

Favier's  notes  on  the  Pratincole  are  confined  to  remarking 
"  that  it  arrives  from  the  south  and  passes  to  Europe  during  the 
month  of  April,  being  observed  returning  thence  in  September 
to  join  those  which  have  remained  near  Tangier  for  the  breeding- 
season.  All  disappear  south  for  the  winter  months." 

We  found  this  bird  in  April,  on  the  dried  mud  at  the  lakes  of 
Meshree  el  Haddar,  south  of  Larache,  in  countless  thousands. 
They  had  not  then  begun  to  lay;  so  possibly  some  of  these 
swarms  would  pass  on  northwards.  We  there  witnessed  a 
number  of  these  birds  mobbing  a  Marsh-Harrier  which  had 
intruded  on  their  ground,  buffeting  and  bullying  him  just  as 
Peewits  will  do  when  a  Hawk  passes  near  their  breeding-ground. 
At  times  at  least  a  hundred  Pratincoles  were  dashing  at  once 
about  the  Hairier,  which  soon  made  its  best  way  out  of  their 
district.  Pratincoles  are  very  crepusculine  in  their  habits, 
flitting  up  and  down  over  the  surface  of  a  river  or  a  pool  much 
after  the  manner  of  the  Indian  Skimmer  (Ehynchops  albicollis) 
very  late  in  the  evening  —  as  late,  indeed,  as  they  can  be 
distinguished.  They  are  then  silent,  but  by  day,  especially 
when  disturbed,  their  cry  is  ceaseless  ;  and  the  Moorish  name 
given  by  Favier  is  doubtless  derived  from,  as  it  is  suggestive  of, 
their  note.  They  are  generally  very  tame  and  fearless,  often 
allowing  one  to  approach  within  a  few  yards,  and  are  birds  of 
very  powerful  flight,  reminding  one  much  of  the  Terns  in  this 
respect. 

On  the  south  side  of  the  Straits  the  Pratincole  is  found  in 
large  numbers  wherever  there  are  lagoons,  which,  drying  up  in 


262  CHARADRIID.E. 

spring,  leave  a  surface  of-  sun-baked  mud  on  which  they  deposit 
their  complement  of  three  eggs  only,  about  the  second  week  in 
May.  The  earliest  egg  I  saw  was  taken  on  the  3rd  of  that 
month,  and  they  mostly  arrive  about  the  20th  of  April,  the 
earliest  date  on  which  I  saw  one  being  the  4th  and  10th  of  that 
month  in  two  consecutive  years.  Verner  noticed  six  at  Palmones, 
near  Gibraltar,  on  the  21st  of  April,  1878.  They  fly  very  high 
when  on  passage,  and  attract  notice  chiefly  from  their  cry.  The 
latest  date  of  the  return  migration  observed  was  the  14th  of 
October,  when  a  young  bird  was  procured. 

In  some  seasons  they  breed  about  the  edges  of  the  Laguna  de 
la  Janda,  but  the  marisma  of  the  Guadalquivir  is  their  chief 
resort. 

A  friend  of  mine,  who  shot  several  on  the  autumnal  passage, 
informed  me  that  they  were  excellent  eating ;  but  in  this  respect 
I  can  give  no  personal  information. 

Axillaries  cJtestnut-red  ;  tail  much  forked. 

Above  dull  brown  ;  rump  and  tail-coverts  white  ;  throat  buff,  bordered 
by  a  black  line  meeting  at  the  gape,  forming  a  bridle  or  collar  ;  wings  very- 
long,  about  7^  inches.  Length  9-10  inches. 


CHARADRIIDJE.     Hind  toe  absent  in  most  species. 

284.  CEdicnenms  scolopax  (S.  G.  Gmelin).    The  Stone-Curlew. 

Moorish.  El  Karuana.     Spanish.  Alcaravan. 

The  Stone-Curlew  is  found  on  both  sides  of  the  Straits  as  a 
resident  in  considerable  numbers,  nesting  generally  about  the 
beginning  of  May,  and  depositing  its  complement  of  two  eggs 
usually  on  stony  dry  ground. 

These  birds  are  far  more  common  in  the  winter  months,  and 
most  so  during  their  migration,  which  is  northwards  during 
March  and  April,  and  southwards  in  October,  November,  and 
December.  They  pass  in  lots  of  from  five  or  six  to  fifty  in 
number,  and  are  chiefly  observed  on  ploughed  fields,  generally 


CURSORIUS  GALLICUS.  263 

near  the  banks  of  rivers,  where  I  have  sometimes'  shot  them  as 
they  flew  by  when  waiting  for  ducks  in  the  evening,  and  they  are 
doubtless  nocturnal  feeding  birds.  Except  in  the  breeding-season, 
when  they  are  very  noisy,  I  have  found  them  rather  wild. 

Two  obscure  white  bands  across  (extended)  wings  ;  breast  streaked  ; 
chin  and  throat  white ;  tail  rather  long  and  graduated ;  bill  yellow, 
terminal  half  black  ;  legs  yellow  ;  iris  very  large  and  yellow.  Length 
1G  inches. 

285.  Cursorius  g-allicus  (J.  F.  Gmelin).    The  Cream-coloured 
Courser. 

Moorish.  Gueta  (Fattier) i  but  this  name  applies  also  to 
Pier  odes  alchata. 

"  This  Courser  appears  annually  during  July  in  some  numbers 
on  the  plains  of  Sharf  el  Akab,  not  far  from  Tangier.  Their  stay 
there  and  their  numbers  vary  according  to  the  abundance  or 
scarcity  of  insects,  and  also  with  the  temperature ;  for  unless  the 
latter  is  favourable,  they  are  rarely  met  with,  and  none  were  seen 
during  the  year  1854.  They  leave  these  plains  in  August  or  the 
first  part  of  September.  Early  in  summer  they  ought  to  be  found 
nesting  near  Sharf  el  Akab,  as  in  May,  1847,  a  male  was  brought 
to  me  by  a  chasseur,  who  rescued  it  from  a  Falcon  which  had 
struck  it  down. 

"Their  food  is  entirely  insects  or  larva?,  particularly Pcntatoma 
torquata  and  different  sorts  of  grasshoppers.  They  are  met 
with  in  small  lots,  usually  frequenting  dry  arid  plains,  where 
they  spread  out  in  all  directions,  running  about  after  insects,  and 
are  very  wary  and  difficult  to  get  a  shot  at.  Their  cry  of 
alarm  is  much  like  that  of  the  Plover.  They  rest  and  sleep 
in  a  sitting  position,  with  their  legs  doubled  up  under  them. 
Should  they  not  fly  away  when  approached  they  run  off  with 
astonishing  swiftness,  manoeuvring  to  get  out  of  sight  behind 
stones  or  clods  of  earth ;  then,  kneeling  down  and  stretching  the 
body  and  head  flat  on  the  ground,  they  endeavour  to  make  them- 


204  CHARADRIID-E. 

selves  invisible, — though  all  the  time  their  eyes  are  fixed  on  the 
object  which  disturbs  them,  and  they  keep  on  the  alert  ready  to 
rush  off  again  if  one  continues  to  approach  them. 

"  The  age  of  the  young  birds  can  be  well  made  out  by  the  zig- 
zag markings  with  which  the  plumage  is  speckled,  which  becomes 
clearer  each  moult  till  the  end  of  the  second  year,  when  they 
assume  the  regular  adult  livery.  There  is  no  difference  at  any 
age  in  the  plumage  of  the  sexes. 

"  In  1849  they  did  not  leave  till  the  llth  of  September,  when 
a  chasseur  brought  me  one  slightly  wounded  in  the  wing.  I 
tried  to  keep  this  bird  alive  ;  but  it  died  directly  the  weather 
became  cold.  It  proved  on  dissection  to  be  a  female  ;  and  from 
the  large  size  of  the  eggs  in  the  ovary  it  appeared  as  if  it  would 
soon  have  nested,  probably  in  October  or  November,  when 
doubtless  they  retire  to  a  much  warmer  climate. 

"Towards  the  end  of  August,  1851,  two  others  were  brought 
to  me,  both  slightly  wounded — one  an  adult,  the  other  an  imma- 
ture bird.  To  prevent  the  birds  this  time  from  dying  of  cold,  I 
placed  them  by  day  in  a  room  where  there  was  always  a  fire  kept 
up.  At  night  I  put  them  in  a  box,  making  a  door  at  the  side, 
lining  the  top  and  sides  with  cotton-wool,  placing  sand  an  inch 
deep  on  the  bottom  ;  this  was  warmed  and  dried  by  putting  a 
charcoal  brazier  inside  during  the  day.  I  fed  the  birds  on  grass- 
hoppers till  November,  when  these  insects  became  very  scarce, 
and,  as  each  bird  ate  fifty  daily,  it  was  necessary  to  change  their 
diet  to  the  larvee  of  coleoptera,  which,  after  some  reluctance,  they 
began  to  take.  This  food  suited  them  better  than  grasshoppers, 
the  birds  becoming  fatter,  at  the  same  time  eating  less.  They 
did  well  till  January,  when,  the  adult  bird  pining  and  refusing 
food,  I  tried  to  save  it  by  cramming ;  but  this  was  useless,  as  it 
died  in  February,  and  on  dissection  I  found  that  death  was 
caused  by  a  very  large  tumour  in  the  stomach.  It  proved  to  be 
a  female ;  and  from  the  ovaries  it  appeared  the  season  for  laying 
had  passed. 


CURSORIUS  GALLICUS.  265 

"The  surviving  bird  continued  well  till  the  end  of  January; 
then,  appearing  ill,  I  fed  it  by  hand  till  April,  when  as  the 
weather  became  warmer  it  grew  more  healthy.  I  then  shut  it 
up  in  a  cage  with  a  white  Turtle-Dove.  The  Courser  was  the 
stronger  bird,  and  did  little  else  than  play  with  the  Dove ;  but 
they  lived  in  perfect  harmony.  In  May,  sexual  desire  was  shown 
in  a  very  marked  manner ;  but,  unluckily,  the  Dove  was  also  a 
female.  During  the  exhibition  of  this  passion  the  Courser  used 
to  make  a  noise  which  may  be  expressed  thus,  '  rerererJ 

"  This  continued  till  the  middle  of  June,  then  entirely  ceasing 
till  the  next  year  (1853),  when  it  resulted  in  the  Courser  laying 
eight  eggs — the  first  on  the  15th,  the  second  on  the  16th,  the 
third  on  the  30th  of  May,  the  fourth  on  the  1st,  the  fifth  on  the 
llth,  the  sixth  on  the  14th,  the  seventh  on  the  23rd,  and  the  last 
on  the  25th  of  June.  In  1854  she  laid  again,  with  the  same 
irregularity,  twelve  eggs — the  first  on  the  17th  of  May,  the  last 
on  the  28th  of  July.  Though  in  perfect  health,  treated  and  fed 
in  the  same  way,  she  did  not  lay  in  1855 — but  in  1856  laid  two 
eggs,  on  the  6th  and  7th  of  July.  In  1857  she  again,  at  irregular 
intervals,  laid  ten  more  eggs — the  first  in  May,  the  last  in  July. 
In  1858  none  were  laid.  In  1859  she  produced  four  more  eggs 
— the  first  two  on  the  6th  and  7th  of  July,  the  others  on  the  9th 
and  10th  of  August. 

"  Shortly  afterwards  this  bird,  in  perfect  health,  plumage,  and 
vigour,  was  lost  to  ornithology,  owing  to  the  war  between  Spain 
and  Morocco  ;  for  on  the  25th  of  October  I  was  ordered,  with 
other  French  subjects,  to  embark  in  the  French  war-steamer 
'  Mouette,'  and  not  knowing  when  I  should  return,  and  still  less 
how  to  take  care  of  my  bird,  I  made  up  my  mind  to  let  it  go  ; 
but  it  was  so  tame  that  it  either  would  not  or  could  not  use  its 
wings ;  so,  in  my  dilemma,  I  gave  it  in  charge  of  a  Moor  during 
my  absence;  but,  unfortunately,  on  my  return  in  April  1860  I 
found  it  had  died. 

"  From  my  observations  it  seems  that  these  birds   could  be 


-66  CHARADRIID.E. 

domesticated  and  bred  so  as  to  be  perhaps  used  for  the  table ;  but 
their  value  would  make  them  rather  expensive  luxuries,  worthy 
of  comparison  with  those  splendid  feasts  given  by  the  ancient 
Roman  Emperors ;  for .  each  bird  would  cost  more  than  twelve 
dozen  capons. 

"  Meanwhile,  considering  the  eggs  laid  by  the  above-named 
female  bird,  the  size  of  the  testes  of  the  males  and  ovaries  of  the 
females  in  August  and  September,  one  may  conclude  that  they 
breed  more  than  once  a  year,  and  that  the  complement  of  eggs  is 
two.  These  are  rather  elliptical  in  shape,  of  a  cream-colour, 
spotted  or  marbled  with  red,  bluish  ash,  and  brown." — Famer. 

This  Courser  occasionally  wanders  across  the  Straits,  having 
been  obtained  near  Malaga,  in  1877,  and  two  which  were  shot 
at  Coria,  near  Seville,  in  1883,  are  in  Lord  Lilford's  collection. 

General  colour  sandy  buff ;  primaries  and  primary-coverts  black ; 
axillaries  Hack',  bill  as  long  as  head,  slightly  curved;  legs  greyish  white. 
Length  9^-10  inches. 

286.  Charadrius  pluvialis,  Linnseus.    The  Golden  Plover. 

Moorish.  Tullit.     Spanish.  Chorlito. 

"  This  Plover  is  very  abundant  around  Tangier  in  large  flocks, 
which  arrive  during  October  and  November,  and  which  return  to 
Europe  in  February  and  March." — Famer. 

The  Golden  Plover  occurs  as  above  on  the  Andalucian  side  of 
the  Straits ;  but  at  the  same  time  their  numbers  fluctuate  very 
much,  in  some  winters  the  quantity  seen  being  very  great.  Upon 
their  first  arrival  they  are  generally  tame ;  but  being  so  much 
sought  after  by  the  cazadores,  they  soon  learn  their  danger  and 
become  more  wary.  The  earliest  noticed  near  Gibraltar  was  on 
the  1st  of  November ;  the  latest  was  on  the  6th  of  March. 

The  best  ground  for  Golden  Plover  is  the  vicinity  of  Tapata- 
nilla.  They  always  frequent  the  same  places  ;  and  if  put  up  from 
any  spot  are  almost  certain  to  return  in  less  than  an  hour.  I 
regret  to  say  that  occasionally  the  Spaniards  catch  them  like 


SQUATAROLA  HELVETICA.  207 

eels,  by  laying  night-lines  or  hooks  baited  with  a  worm  in  their 
feeding-places :  this  is  a  most  cruel  method  of  procuring  them, 
as  the  unfortunate  birds  linger  in  agony  for  hours,  often  being 
left  till  they  flutter  themselves  to  death. 

Male  in  spring.  Above  blackish,  spotted  with  yellow;  chin,  throat,  breast, 
and  belly  black. 

Female.   Less  black  below,  being  mottled  with  white  feathers. 

In  winter.  Below  white,  chest  and  flanks  marked  with  brown.  Axilluries 
white.  Lenoih  9  inches. 

o 

287.  Charadrius  fulvus,  J.  F.  Gmelin.    Asiatic  Golden  Plover. 

A  specimen  of  this  Eastern  Golden  Plover  was  killed  near 
Malaga  in  1878,  and  is  in  Lord  Lilford's  collection. 
Axillaries  smoky  grey;  otherwise  as  last,  but  slightly  smaller. 

288.  Squatarola  helvetica  (Liunseus).    The  Grey  Plover. 

Spanish.  Redolin. 

Favier  only  remarks  that  "  this  Plover  is  found  near  Tangier 
between  the  months  of  December  and  March." 

The  Grey  Plover  appears  chiefly  to  arrive  near  Gibraltar  during 
the  middle  of  November,  and,  though  frequently  seen  in  autumn 
and  spring,  cannot  be  said  to  be  at  any  time  abundant.  I  have 
noticed  them  on  the  small  plash  of  water  which  after  heavy  rains 
is  formed  on  the  western  part  of  the  Neutral  ground. 

On  the  22nd  of  May,  1869,  I  killed  a  pair  at  one  shot  near 
the  mouth  of  the  Guadiaro,  the  male  being  in  almost  perfect 
summer  plumage,  the  female  not  being  so  far  advanced,  and  the 
eggs  in  her  ovaries  very  slightly  developed.  It  is  very  remarkable 
that  this  northern-breeding  bird  should  linger  so  late  in  such  a 
sunny  southern  country ;  and  the  day  above  mentioned  was  very 
hot  for  the  time  of  the  year.  But  the  Grey  Plover  is  not  singular 
in  thus  remaining  south  so  late ;  both  the  Knot  and  the  Curie \v 
Sandpiper  loiter  equally  late  into  the  spring. 

The  marks  by  which  this   species  is  to  be  distinguished  from 
the  Golden  Plover  are  the  presence  of  a  very  small  hind  toe  and 


2G8  CIIAKADmiD.E. 

the  black  axillaries ;  it  is,  moreover,  a  larger  bird,  and  shore- 
frequenting,  very  rarely  being  found  inland  like  the  Golden 
Plover. 

Axillaries  black  ;  small  hind  toe  present. 

In  spring.  Above  white,  barred  with  black  ;  underparts  black,  or  black 
marked  with  white. 

In  winter  marked  much  as  Golden  Plover,  but  the  yellow  spots  of  that 
species  are  replaced  by  white,  young  birds  only  being  marked  with  a  few 
yellow  spots.  Length  10-12  inches. 

289.  Eudromias  morinellus  (Linnaeus).    The  Dotterel. 

"  This  bird,  which  appears  to  travel  in  company  with  Cursorius 
isabellimts,  is  found  near  Tangier  sparingly  on  its  annual  passage 
during  August  and  September,  frequenting  in  small  flocks  the 
same  dry  places  that  the  Courser  inhabits ;  and,  like  them,  it 
seems  to  dread  the  cold." — Favier. 

Could  Favier  occasionally  have  seen  the  Dotterel  on  mountain- 
tops  in  Scotland,  he  would  not  have  supposed  them  to  fear  the 
cold  ;  but  curiously  enough  he  omits  to  mention  the  date  of  their 
appearance  in  spring ;  and  I  have  no  record  in  my  notes  of 
observing  them  at  that  season  on  the  Spanish  side  of  the  Straits  : 
the  few  I  have  seen  were  in  autumn,  the  latest  being  shot  about 
the  9th  of  November.  Probably  they  pass  straight  on,  and  thus 
appear  rarer  than  is  the  case ;  but  doubtless  their  line  of  migra- 
tion must  be  further  to  the  east. 

Axillaries  greyish.  Crown  nearly  black  with  white  lines  below,  all  round 
above  eyes  to  bill;  chin  and  throat  white;  upper  breast  chestnut-red;  belly 
black. 

Females  brighter  than  males.     Length  9  inches. 

290.  JEglalitis  hiaticula  (Linnseus).    The  Ringed  Plover. 

Moorish.  Kouba  (Drummond-Hay)  ("  the  hooded  one,"  vide 
Crested  Lark).  Spanish.  Andarios,  Correrios,  Frailecillo  (the 
first  two  applied  to  all  small  Waders). 

"  This  Ringed  Plover  is,  near  Tangier,  found  in  small  numbers 
in  pairs  and  companies  on  the  sea-shore.  They  arrive  during  the 


uEGIALITIS  CURCmCA.  2C9 

months  of  September,  October,  and  November,  returning  north 
again  in  April  and  May." — Favier. 

Though  without  absolute  proof,  I  am  nearly  sure  that  this 
species  occasionally  remains  to  breed  near  Gibraltar,  having  shot 
them  as  late  as  the  28th  of  May,  and  having  seen  eggs  obtained 
near  Seville  as  early  as  on  the  23rd  of  March ;  but  this  is  the 
only  instance  I  know  of  their  nesting  so  far  south.  During 
autumn  and  until  April  the  Ringed  Plover  is  extremely  plentiful 
along  the  coast,  and  most  so  in  the  month  of  March. 

Above  hair-brown  ;  forehead  white  ;  fore  crown  black,  reaching  to  and 
below  eyes  on  each  side,  and  to  base  of  bill  and  ear-coverts  ;  throat  white, 
forming  a  complete  ring,  below  this  a  broad  black  band  round  breast ;  both 
these  collars  narrowing  to  a  line  round  nape  ;  rest  of  underparts  white  ; 
eyelids,  legs,  and  feet  orange  ;  basal  half  of  bill  yellow,  rest  black. 

Females  have  the  collars  less  clearly  defined. 

Young  brown  where  black  in  adults  ;  bill  black  ;  legs  brown.  Length 
7-7^  inches. 

291.  ^gialitis  curonica  (J.  F.  Gmelln).    The  Little  Ringed 
Plover. 

This  small  inland  species  is  not  mentioned  by  Favier  in  his 
MS.,  although  included  by  Mr.  Drake  in  his  '  List  of  the  Birds 
of  Morocco.'  We  found  them  very  common  in  that  country  on 
river-banks  during  the  month  of  April.  Equally  abundant  at 
that  time  on  the  Spanish  side,  the  Little  Ringed  Plover  is  only 
seen  during  the  breeding-season.  They  mostly  arrive  about  the 
14th  of  March,  some  passing  on ;  others  remain  to  nest, 
depositing,  about  the  14th  of  May,  four  eggs  on  the  sand  or 
shingle  by  the  sides  of  rivers.  The  earliest  eggs  Major  Verner 
found  were  on  the  loth  May,  1875.  Many  pairs  nest  on  the 
river  Barbate,  near  Casas  Viejas.  There  is,  of  course,  no 
approach  to  a  nest ;  but  the  eggs,  with  the  small  ends  inwards, 
are  placed  in  a  slight  depression  probably  formed  by  the  birds 
themselves  in  the  sand  or  gravel. 

Resembles  last,  but  much  smaller,  is  a  more  inland  bird,  and  has  shaft  of 
first  primary  only  white  ;  eyelids  golden  yellow  ;  legs  and  feet  pale  brown. 
Length  6£  inches. 


270  CIIARADRIID.E. 

292.  ^Eg-ialitis  cantiana  (Latham).    The  Kentish  Plover. 

Moorish.  Bou-hejaira  (father  of  stones,  Fauier}.  Spanish. 
Charran,  Andarios. 

"  This  bird  is  very  Abundant  near  Tangier,  and  generally  found 
at  the  mouths  of  rivers.  Many  are  resident,  those  which  are 
migratory  arriving  during  September  and  October,  leaving  north- 
wards in  March  and  April." — Favier. 

The  Kentish  Plover  is  by  far  the  most  plentiful  of  the  sea- 
shore Waders  on  the  Spanish  side ;  and  they  are  always  very 
tame,  being  seldom  molested  by  the  Spaniards.  The  local  name 
of  Charran,  which  I  have  heard  for  them  near  Gibraltar,  signifies 
a  low  unmitigated  blackguard,  and  the  application  of  which  to 
this  harmless  and  charming  bird  is  not  easy  to  understand.  This 
name  is  also  applied  near  Malaga  to  some  species  of  Terns. 

This  little  Plover  is  found  throughout  the  year,  but  is  most 
abundant  during  the  seasons  of  migration  ;  they  are  very  active, 
nimble  little  birds,  running  along  the  shore  sometimes  in  front 
and  within  a  yard  of  one's  horse's  feet ;  frequently  running  out 
on  the  wet  sand  as  one  wave  recedes,  to  rush  back  again  as 
another  returns,  like  the  Sanderling ;  but  they  are  much  too 
agile  ever  to  be  caught  by  the  influx. 

About  the  20th  of  April  they  commence  to  lay  their  four  stone - 
coloured  eggs,  marked  with  black  spots  and  streaks.  Some  regu- 
larly breed  on  the  dry  sandy  hillocks  and  banks  near  the  mouth 
of  the  First  River  (Guadarranque) ;  but,  like  the  other  species  of 
jEgialitis,  they  frequently  nest  far  away  from  the  shore,  as  on  the 
dried  mud  of  the  marisma.  Verner  took  four  fresh  eggs  on  the 
9th  of  May,  1879,  near  Gibraltar,  and  he  remarks  that  the 
Kentish  Plover  in  England  nearly  always  lays  only  three  eggs ; 
that  is  to  say,  between  1885  and  1893,  of  many  nests  he  saw 
only  one  contained  four  eggs,  and  the  fishermen  about  Dungeness, 
who  rob  the  nests  for  collectors,  assert  that  three  eggs  and  no 
more  is  the  invariable  number. 

Bill  and  legs  black. 


VANELLUS  VULGARIS.  271 

Male.  Above  hair-brown  ;  forehead  white,  black  patch '  on  fore  crown  ; 
rest  of  crown  and  nape  buff ;  black  line  through  eye  from  bill  to  ear- 
coverts  ;  a  black  patch  on  each  side  of  breast,  not  forming  a  collar  ;  rest  of 
underparts  white. 

Female.  No  black  on  fore  crown,  but  that  and  crown  uniform  with  the 
back  ;  the  patch  on  each  side  of  breast  same  colour  as  back.  Length 
6  inches. 


293.  Vanellus  vulgaris  *,  Bechstein.    The  Peewit,  Lapwing, 
or  Green  Plover. 

Moorish.  Biliat,  El  Thudi  (the  Jew).  Spanish.  Ave  fria  (cold 
bird),  Judia  (Jewess). 

"  This  Plover  occurs  near  Tangier  in  abundant  flocks  throughout 
the  winter  months,  arriving  from  the  north  during  October  and 
November,  crossing  back  again  to  Europe  in  February  and 
March. 

"The  superstitious  Arabs  believe  that  these  birds  are  Jews 
changed  into  the  shape  of  birds,  and  also  believe  that  they  still 
retain  all  their  Israelitish  characteristics,  even  wearing  the  black 
Hebrew  skull-cap." — Favier.  Hence  the  name  "  El  Thudi." 

We  observed  towards  the  end  of  April  three  or  four  pairs  of 
Peewits,  which  were  nesting  at  the  northern  end  of  the  lakes  of 
Ras  el  Doura.  As  we  had  not  sufficient  time  to  go  further  than 
the  commencement  of  these  lakes,  possibly  many  others  were  to 
be  found  breeding  still  further  south.  The  place  where  we  saw 
them  was  some  eighty  miles  at  the  least  to  the  south  of  Tangier. 
On  the  Spanish  side  of  the  Straits  very  few,  compared  with  their 
numbers  in  winter,  remain  to  breed  in  the  marismas  of  the 
Guadalquivir,  where  we  found  the  nest  with  young  on  the  26th 
of  April.  Curiously,  none  remain  to  breed  about  the  Laguna  de 
la  Janda,  or,  as  far  as  could  be  ascertained,  anywhere  but  in  the 
marisma.  The  majority  of  the  Peewits  arrive  near  Gibraltar 

*  This  is  an  instance  of  abolishing  the  long-used  and  distinctive  synonym 
(cristatug)  for  a  vague  one,  because  of  two  years'  priority. 


272  CHARADKIID^E. 

about  the  middle  of  October,  and  take  their  departure  north 
about  the  first  week  in  March.  During  the  winter  they  are  to  be 
found  on  every  level  piece  of  ground  ;  and  have  occasionally  been 
seen  on  the  green  glacis  of  the  batteries  near  the  Alameda  at 
Gibraltar,  and  often  noticed  on  the  "North  Front";  while  at 
times  they  are  found  on  hill-sides  at  a  considerable  elevation  if 
there  be  any  grassy  and  suitable  open  spot.  They  seem  to  be 
more  scattered  and  dispersed  about  than  is  usual  in  England, 
although  large  flocks  may  sometimes  be  seen. 

The  Peewit  used  to  be  an  unfailing  source  of  diversion  to  the 
British  subaltern,  and  also  to  the  "  sportsman  "  of  the  Rock,  who, 
a  marvel  of  leather  straps,  gaiters,  bags — leather  all  over — used 
to  sally  out  of  Gibraltar  for  a  Sunday's  shooting,  accompanied 
generally  by  a  bob-tailed,  mangy,  lean,  and  hungry-looking 
species  of  pointer — by  appearance  warranted  to  devour  imme- 
diately anything  his  master  might  kill ;  but  the  unfortunate 
animal  probably  seldom  had  the  opportunity  of  having  its 
appetite  so  gratified. 

Hind  toe  present  and  well  developed. 

Above  glossy  green ;  crown,  throat,  and  breast  black  ;  crest  of  long 
slender  black  feathers  ;  upper  and  under  tail-coverts  rufous  ;  legs  flesh- 
coloured. 

Young.  Crest  short ;  feathers  above  with  buff  edges  ;  legs  dark  brown. 
Length  13  inches. 

294.     Vanellus     gregrarius     (Pallas).      The     Black-bellied 
Lapwing1. 

Mr.  Saunders  mentions  having  seen  one  specimen  of  this 
Lapwing  hanging  up  in  Cadiz  market,  whilst  Arevalo  records 
some  from  near  Malaga. 

Is  an  eastern  bird,  having  the  gregarious  habits  of  the  Golden 
Plover.  When  on  the  wing  the  white  secondaries  show 
markedly. 

Secondaries  entirely  white  ;  shortest  primaries  black  on  the  outer  webs 
and  white  on  the  inner  webs  ;  hind  toe  present.  Length  13  inches. 


STREPSILAS  INTERrRES.  273 

Adult.  Crown  black,  bordered  below  by  a  broad  white  stripe  above  eye 
to  nape  ;  belly  black,  bordered  below  with  chestnut  to  the  vent. 

Young.  Crown  dark  brown  ;  band  from  forehead  to  nape  buffish  white  ; 
breast  white,  marked  with  dark  brown. 


295.  Hsematopus  ostralegus,  Linnaeus.    The  Oystercatcher. 

Moorish.  Aisha  el  behar  (Favier)  (the  pied  one  of  the  sea) 
Spanish.  Ostrero  (Arevalo). 

"  This  species  is  found  near  Tangier  on  passage,  passing  north 
during  April  and  May,  and  returning  in  October." — Favier. 

Favier  also  states  that  the  Oystercatchers  sometimes  remain  to 
nest.  Very  possibly  this  is  the  case ;  but  the  eggs  which  were 
marked  as  "  Oystercatcher's  "  by  him  were  to  all  appearance  those 
of  the  Stone-Curlew. 

On  the  Spanish  side  this  bird  is  not  at  all  numerous  near 
Gibraltar,  and  appears  irregularly  from  autumn  to  spring,  the 
latest  recorded  was  one  observed  by  Lord  Lilford  on  the  5th  of 
May  near  the  mouth  of  the  Guadalquivir. 

No  hind  toe  ;  bill  yellow,  longer  than  head. 

Plumage  black  and  white  ;  head,  neck,  scapulars,  and  terminal  half  of 
tail  black ;  rump,  upper  tail-coverts,  and  axillaries  white ;  legs  pink  ; 
eyelids  crimson.  Length  16  inches. 

296.  Strepsilas  interpres  (Linnaeus).    The  Turnstone. 

Moorish.  Shorno  (Favier).  Spanish.  Revuelve-piedras  (Malaga) 
(Arevalo). 

"  This  bird  is  not  numerous,  being  found  near  Tangier  on  the 
coast  and  sometimes  on  the  edges  of  freshwater  lakes.  They  are 
more  abundant  near  Rabat.  Arriving  from  the  north  in 
September,  they  return  in  February." — Favier. 

The  Turnstone  is  found  on  the  Spanish  coast  in  autumn  and 
spring.  I  never  saw  them  in  any  numbers,  and  chiefly  observed 
them  about  the  end  of  March,  occasionally  in  company  with  the 
Ringed  Plover. 


274 

Hind  toe  present ;  bill  black,  strong,  straight,  and  pointed,  shorter  than 
head  ;  chin  and  throat  pure  white. 

In  spring.  Back  and  wings  marked  with  chestnut  and  black  ;  loicer  back 
and  rump  ichite,  the  latter  crossed  by  a  black  band  ;  legs  and  feet  orange- 
red. 

In  winter,  chestnut  colour  absent  ;  legs  pale  red.     Length  9^  inches. 


SCOLOPACIDJE. 

297.  Recurvirostra  avocetta,  Linnaeus.    The  Avocet. 

Moorish.  Bou-am-hait.     Spanish.   Avoceta. 

"  This  species  is  not  common  in  the  vicinity  of  Tangier,  being 
only  met  with  on  passage,  on  the  edges  of  rivers  and  lakes,  in 
small  flights,  which  pass  northwards  during  March,  April,  and 
May  and  return  south  in  November." — Favier. 

I  never  met  with  the  Avocet  near  Gibraltar.  In  some  seasons 
many  nest  in  some  parts  of  the  marismas  during  the  month  of 
May,  and  specimens  of  both  eggs  and  birds  are  occasionally 
brought  into  Seville. 

Bill  black,  curved  upwards  and  pointed. 

White,  except  crown  to  below  eye,  hind  neck,  primaries,  and  some  of 
the  inner  secondaries  and  wing-coverts,  which  are  black  ;  legs  and  the 
partially  webbed  feet  light  blue  ;  iris  red. 

Young.  Crown  and  hind  neck  brown.     Length  18  inches. 

298.  Himantopus  candidus,  Bonnaterre.     The  Black-winged 

Stilt. 

Moorish.  Bou-ksaiba  (Favier)  *.     Spanish.  Cigiienela. 

"  This  bird  is  not  found  close  to  Tangier,  but  frequents  fresh- 
water lakes  further  south,  where  many  remain  for  the  breeding- 
saason;  others,  arriving  during  the  month  of  April,  pass  on 
northwards  and  return  in  November.  They  appear  to  migrate 
by  night." — Favier. 

*  "  Ksaiba  "  is  diminutive  of  "  Kaaba,"  a  reed,  and  probabty  applies  to  the  thin 
legs  of  this  Stilt,  not  to  their  aquatic  habits. 


PIIALAROPUS  FULICARIUS.  275 

This  Stilt  is,  in  spring,  one  of  the  most  common  of  the  marsh- 
birds  on  both  sides  of  the  Straits.  At  Meshree  el  Haddar,  in 
Morocco,  and  in  the  marisma  of  the  Guadalquivir  their  numbers 
were  perfectly  marvellous.  In  some  seasons  they  have  nested  at 
the  Laguna  de  la  Janda.  They  frequent  open  shallow  pools  and 
lakes,  and  are  very  seldom  seen  where  there  is  grass  or  rushes, 
being,  as  a  rule,  very  tame  and  confiding ;  while  their 
conspicuous  black-and-white  plumage  and  noisy  habits  render 
them  certain  to  attract  attention,  either  as  they  fly  with  their 
long  pink  legs  stretched  out,  Heron-like,  behind  them,  or  as  they 
wade  about,  usually  up  to  their  knees,  in  the  shallow  water, 
where  they  seek  their  food  in  the  shape  of  aquatic  insects,  gnats, 
and  flies. 

The  Black-winged  Stilt  is  almost  entirely  migratory ;  but  in 
some  years  a  few  undoubtedly  remain  in  Spain  throughout  the 
winter,  as  I  have  seen  small  lots  on  the  26th  and  27th  of 
November  in  different  years,  many  on  the  22nd  of  December, 
and  others  on  the  14th  of  January.  The  chief  number  appear 
towards  the  end  of  March  and  beginning  of  April ;  and  they  are 
then  not  (infrequently  seen  near  Gibraltar  at  the  mouths  of  various 
rivers,  but  soon  pass  on  to  their  breeding-places,  where  they  nest 
in  colonies  and  deposit  their  four  eggs  on  the  half-dried  mud. 
I  have  seen  eggs  as  early  as  the  28th  of  April;  but  the  majority 
lay  about  the  10th  of  May. 

Bill  black,  twice  as  long  as  head,  slender,  straight,  and  ointed  ;  hind 
toe  absent ;  nape,  upper  hind  neck,  wings,  and  back  black  ;  underparts 
white  ;  legs  pale  pink,  about  4^  inches  long. 

Old  males  have  pure  white  heads.     Length  13  inches. 

299.  Phalaropus    fulicarius    (Linnaeus).      The  Lobed-footed 
Stint  or  Grey  Phalarope  *. 

Favier  mentions  only  two  specimens  of  this  bird  as  having  been 

*  "  Phalarope,"  like  "  Hemipode,"  is  an  Anglicized  compound  name  only- 
understood  by  ornithologists.  An  English  name,  as  used  by  Jerdon,  is  more 
intelligible  to  the  world  in  general. 

T2 


276 

obtained  by  him  near  Tangier,  in  December,  1858.  Mr.  Drake 
also  refers  to  this  Phalarope  as  having  been  shot  near  Tangier 
during  the  month  of  January ;  and  others  have  since  occurred. 

On  the  Spanish  side  one  was  killed  on  29th  of  November, 
1872,  others  on  the  4th  of  December,  1875,  and  they  occur 
irregularly,  as  in  England,  sometimes  in  numbers. 

These  birds  never  appear  to  go  on  the  land  except  to  nest,  and 
are  always  seen  swimming,  often  far  at  sea. 

Front  toes  as  in  next  species  lobed,  like  a  Coot. 
Bill  short  and  wide  ;  tail  graduated. 
In  summer.  Below  chestnut. 

In  winter.  Forehead  and  crown  white ;  grey  nape,  white  wing-bar  f 
below  white.  Length  8  inches. 

300.  Scolopax  rusticula,  Linnaeus.    The  Woodcock. 

Moorish.  Hejel  el  himar  (red  partridge),  Khadim  el  ajel  (the 
servant  of  haste),  Bou-monkar  (Father  of  the  bill),  applied  to  all 
long-billed  birds.  Andalucian.  Gallineta.  Spanish.  Chocha. 

The  Woodcock,  according  to  Favier,  is  "  not  abundant  around 
Tangier,  arriving  during  November  and  departing  in  March." 

Uncertain,  both  in  numbers  and  as  to  time  of  arrival  near 
Gibraltar,  in  some  seasons  Woodcocks  are  tolerably  plentiful,  as 
in  1873  ;  in  others,  as  in  the  winter  of  1871-72,  they  are  very 
scarce.  Five  or  six  couple  in  the  day  for  two  guns  is  a  very  fair 
bag ;  but  there  was  an  instance  of  a  Spanish  cazador  bagging 
twenty-one  in  a  day  near  Algeciraz ;  those  who  wish  for  good 
Woodcock  -  shooting  had  better  not  try  either  Andalucia  or 
Morocco,  but  go  to  the  east  of  the  Mediterranean. 

My  earliest  note  of  the  arrival  of  a  Woodcock  about  Gibraltar 
was  on  the  17th  of  October,  but  very  few  arrive  until  the  middle 
of  November.  The  latest  noticed  was  on  the  8th  of  March ;  but 
I  have  seen  them  in  Seville  market  on  the  22nd  of  that  month. 
I  obtained  near  that  city  a  fine  white  variety,  which  is  now  in 
the  Norwich  Museum. 

Second-hand  information  is  not  always  trustworthy,  but  the 


GALLLYAGO  MAJOR.  277 

postmaster  at  San  Roque,  the  late  Mr.  Macrae,  an  official  well 
known  in  those  clays  to  the  garrison  of  Gibraltar,  and  upon  whose 
veracity  and  knowledge  of  the  bird  I  can  depend,  told  me  that 
once,  and  only  once,  he  saw  at  break  of  day  a  regular  flight,  or 
what  the  Spaniards  would  call  a  "  band  "  of  Woodcocks  passing 
south.  He  described  them  as  being  about  twenty  or  thirty  in 
number,  but  the  light  was  so  dim  he  could  not  see  where  they 
went  to. 

Feathers  on  breast  barred  ;  sixteen  tail-feathers  with  silvery-white  tips 
on  underside  ;  greyish  tips  above.  Length  13  inches. 

301.  Gallinag-o  major  (J.  F.  Gmelin).    The  Great  or  Solitary 
Snipe. 

Spanish.  Agachadiza  real. 

Favier  only  mentions  a  single  specimen  of  this  Snipe  as  having 
been  obtained  by  him  near  Tangier,  in  1859.  It  is,  however, 
included  in  Mr.  Drake's  list,  '  Ibis,'  1869,  p.  153,  as  twice  noticed 
in  March. 

The  Great  Snipe  is  only  met  with  near  Gibraltar  on  passage, 
"  here  to-day,  gone  to-morrow."  I  saw  two  and  shot  one  at 
Casas  Viejas  on  the  24th  of  October,  1868  ;  one  was  killed  near 
Gibraltar  on  the  17th  of  October,  1871 ;  and  I  know  of  another 
obtained  in  April.  It  is  there  a  well-known  bird,  but,  passing 
north  late  in  April  and  early  in  May,  and  returning  again  in 
September  and  October,  is  not  very  liable  to  be  noticed;  and 
probably  their  chief  line  of  migration  lies  more  to  the  eastward. 

This  Snipe  is  usually  very  tame,  and,  lying  closely,  shows  the 
external  white  feathers  of  the  tail  very  much  when  rising ;  and 
generally  alights  again  within  a  short  distance,  never  uttering 
any  sound. 

Belly  barred  to  the  vent  ;  sixteen  tail-feathers,  lower  half  of  the  four 
outside  pairs  white  in  adults,  white  with  brown  bar  in  young.     Length 
inches. 


278  SCOLOPACID.E. 

302.  Gallinag-o  coelestis  (Frenzel).    Tlie  Common  Snipe. 

Moorish.  Bon  monkar  (Father  of  the  bill).  Spanish.  Near 
Gibraltar,  Agachadiza ;  further  north,  Agachona ;  near  Malaga, 
Agachadera. 

Favier  remarks  that  on  the  African  side  "  the  Common  Snipe 
is  fonnd  very  plentiful  around  Tangier  from  the  month  of  October 
until  February,"  which  may  be  said  of  them  likewise  as  regards 
the  Spanish  side  of  the  Straits ;  and  although  better  sport  is  to 
be  had  with  this  (in  a  sporting  sense)  king  of  birds  on  the 
Moorish  side,  the  amusement  is,  as  has  been  already  stated, 
greatly  reduced  by  the  want  of  accommodation  and  utter  absence 
of  comfort ;  not  that  there  is  much  of  the  latter  in  many  places 
on  the  Andalucian  side.  At  Casas  Viejas,  Snipe  sometimes 
arrive  as  early  as  the  beginning  of  September.  I  have  heard  of 
a  straggler  during  August  (one  was  shot  by  Verner  in  the  Soto 
Malabrigo  on  the  21st  of  that  month  in  1879),  but  the  greater 
quantity  do  not  put  in  an  appearance  till  the  end  of  October  and 
the  first  week  in  November.  They  commence  their  departure 
in  March ;  and  by  the  first  week  in  April  all  have  disappeared 
except  a  stray  loiterer,  perhaps  a  wounded  bird.  We  noticed 
one  as  late  as  the  3rd  of  May,  having  observed  it  for  several  days 
previously  in  the  same  situation,  and  would  not  shoot  it,  wishing 
to  see  how  long  it  would  remain  :  this  bird  did  not  appear  to 
have  anything  the  matter  with  it.  I  never  heard  the  drumming 
noise  of  the  Snipe  in  Andulacia — though  at  home  in  England  I 
have  occasionally  heard  them  drumming  of  an  evening  in  the 
New  Forest  as  early  as  the  20th  of  January,  the  weather  then 
being  unusually  mild,  and  the  place  where  they  were  heard 
being  their  regular  nesting-ground. 

I  have  often  noticed  that,  in  the  marshes  both  in  Morocco  and 
Andalucia,  the  best  ground  for  Snipe  was  a  spot  where  sedges 
and  rushes  had  been  burnt  during  the  summer ;  but  the  conse- 
quent absence  of  cover  in  these  places  rendered  it  useless  to  try 
and  walk  up  to  the  birds,  and  the  only  way  was  to  stand  or  sit 


O 

h- 
O 
CO 

O 
O 

QL 

m 


GALLIXAGO  GALLINULA.  279 

perfectly  still  in  the  most  favourite  spot  and  await  their  return. 
I  have  more  than  once  taken  a  chair  down  and  sat  in  it,  waiting 
for  their  flight  overhead,  much  to  the  astonishment  of  the  native 
population,  who  could  not  understand  such  a  proceeding. 

Varieties  or  races  of  this  species,  varying  in  size  and  colour, 
have  been  named  russata,  delamottii,  and  brehmi;  but  it  is 
difficult  to  perceive  any  distinction,  as  these  varieties  are  not 
constant.  "  Sabine's  Snipe  "  is  merely  a  dark  variety. 

Axillaries  white,  barred  with  blackish  grey ;  tail-feathers  fourteen. 
Length  10^  inches. 

303.  Gallinago  gallinula  (Linnaeus).    The  Jacksnipe. 

Moorish.  Saiga  (Favier). 

This  bird  is  stated  by  Favier  to  be  "  nearly  as  common  in  the 
winter  months  around  Tangier  as  the  Common  Snipe,  arriving 
during  November,  and  departing  northwards  in  February." 

On  the  Spanish  side  of  the  Straits  the  Jacksnipe  is  generally 
distributed  throughout  the  winter,  and  is  extremely  numerous 
about  some  favourite  black  muddy  spots  at  Casas  Yiejas  and  in 
the  "  ojos,"  or  land-springs,  at  the  edges  of  the  marisma,  but  is 
by  no  means  so  plentiful  as  the  Common  Snipe.  Towards  the 
end  of  February,  Jacksnipes  assemble  together  very  much ;  and 
this  gathering  of  them  is  a  sure  prelude  to  the  general  departure 
of  most  of  the  Snipes  for  the  north.  The  greatest  number  of  the 
present  species  that  I  ever  saw  anywhere  was  in  some  of  the 
"  ojos  "  westward  of  Coria  del  Rio,  near  Seville ;  these  circular 
spots,  about  ten  yards  in  diameter,  were  very  muddy  and  sparingly 
covered  with  short  sedge.  Many  of  them  held  fifteen  or  a  dozen 
Jacksnipe  ;  and  the  often-cited,  and  perhaps  imaginary,  individual 
who  is  said  to  have  found  a  single  Jacksnipe  afford  him  sport  for 
months  until  his  friend  unluckily  killed  it,  would  here,  indeed, 
have  been  in  happy  hunting-grounds. 

I  could  not  ascertain  any  good  local  Andalucian  name  for  this 
bird.  I  have  heard  some ;  but  they  were  too  absurd  and  varied 


280  SCOLOPACID^E. 

to  repeat.  The  Jacksnipe  is  said  occasionally  to  arrive  in  Anda- 
lucia  towards  the  end  of  September ;  but  my  two  earliest  notes  of 
their  arrival  are  the  24th  and  the  27th  of  October  near  Seville. 

Purple  gloss  on  back  ;  axillaries  white  ;  tail-feathers  twelve,  centre  pair 
longer  than  the  rest.  Length  1\  inches. 

304.  Calidris  arenaria  (Linnaeus).    The  Sanderling1. 
Spanish.  Churrilla  de  tres  dedos  (Malaga)  (Arevalo). 

"  This  bird  is  abundant  during  migration  near  Tangier  in  small 
flocks  along  the  coast,  crossing  the  Straits  during  March,  April, 
and  May ;  they  are  found  returning  south  as  late  as  December. 
I  found  numbers  near  Tetuan  in  February  1848  at  the  mouth  of 
the  river,  where  they  are  known  to  the  Moors  under  the  name  of 
Medrouan." — Favier. 

On  the  African  side  we  saw  large  flights  of  Sanderlings  early 
in  April  between  Tetuan  and  Ceuta.  On  the  Spanish  side  they 
are  common  from  autumn  to  spring  along  the  sea-shore ;  the 
latest  noticed  was  during  the  first  week  in  May. 

Legs  and  bill  black  ;  bill  as  long  as  the  head  and  straight;  no  hind  toe. 

In  spring.  Upper  parts  rufous  and  black;  head,  throat,  and  upper  breast 
rufous,  marked  with  black. 

In  winter.  Grey  above,  white  below  ;  rump  ashy  grey ;  bill  as  long  as 
the  head.  Length  8  inches. 

305.  Tring-a  subarquata,  Giildenstadt.     The    Curlew    Sand- 

piper. 

All  that  Favier  has  in  his  MS.  relative  to  this  bird  is  that  it 
"  passes  near  Tangier  during  the  month  of  April,  returning  south 
in  September." 

I  never  obtained  the  Curlew  Sandpiper  on  the  autumnal 
passage ;  but  in  some  years  vast  numbers  passed  at  Gibraltar 
towards  the  end  of  April,  usually  in  lots  of  from  ten  to  twenty 
in  number  ;  they  were  occasionally  mixed  up  with  Dunlins,  and 
were  chiefly  to  be  seen  at  the  mouths  of  rivers,  particularly  about 
Palmones,  where  one  was  shot  with  the  first  primary  in  each  wing 


TRINGA  ALPINA.  281 

white.  When  flying  they  are  easily  distinguished  by  the  white 
rump,  which  is  then  very  conspicuous.  They  are  in  good  red  or 
breeding-plumage  by  the  26th  of  April ;  that  is  to  say,  the  male 
birds  are ;  but  the  females  are  slower  in  assuming  this  dress,  and 
probably  never  become  as  bright  as  their  mates.  About  Gibraltar 
this  Sandpiper  and  others  bear  the  trivial  name  "  pitillo."  Lord 
Lilford  informs  me  that  he  met  with  the  present  species  at  the 
same  place  and  time  as  the  Knots,  and  in  equal  numbers. 
Curiously,  during  that  spring,  Curlew  Sandpipers  were  unusually 
abundant  near  Gibraltar,  but  not  a  single  Knot  did  I  obtain  or 
see.  There  is,  however,  not  very  much  ground  suitable  for  the 
various  species  of  Tringa  in  the  vicinity  of  the  Kock. 

Bill  curved  downwards  ;  rump  white. 

In  spring.  General  colour  dark  reddish  chestnut,  marked  above  with 
blackish. 

In  autumn.  General  colour  grey  above;  below  white.     Length  7  inches. 

306.  Tringa  canutus,  Linnaeus.    The  Knot. 

Favier  merely  remarks  of  this  species  that  they  "  pass  near 
Tangier  in  June."  If  such  be  the  case,  it  must  be  very  early  in 
that  month.  The  Knot  is  somewhat  irregular  in  appearance 
about  the  vicinity  of  Gibraltar ;  and  I  have  rarely  met  wdth  any, 
and  the  few  seen  have  only  occurred  in  April  and  May.  Lord 
Lilford,  however,  found  them  in  countless  numbers  about  the 
10th  of  May  near  the  edge  of  the  Goto  de  Donana.  They  were 
at  that  time  in  their  fine  red  or  summer  plumage,  and  doubtless 
en  route  to  their  breeding-grounds  in  the  extreme  north. 

Bill  black,  quite  straight,  longer  than  head. 

Plumage  changes  much  as  in  last,  but  the  red  is  not  so  dark  ;  rump  and 
upper  tail-coverts  whitish,  barred  with  blackish  ;  legs  and  feet  black. 
Length  10  inches. 

Young.  Feathers  on  back  bordered  with  dark  bars,  tipped  with  buffy 
white  ;  below  tinged  with  buff;  legs  and  feet  yellowish  green. 

307.  Tringa  alpina,  Linnaeus.    The  Dunlin. 

Favier  remarks  that  the  Dunlin  "  passes  to  Europe  from  the 


SCOLOPACID.E. 

Moorish  coast  during  the  months  of  April,  May,  and  June,  re- 
turning to  winter  further  south  in  October  and  November." 

This  well-known  species,  however,  is  to  be  seen  throughout 
the  winter  near  Gibraltar,  sometimes  in  considerable  numbers. 
Occasionally  they  wander  far  up  the  rivers  some  distance  from 
the  sea,  especially  in  the  spring. 

The  majority  of  these  Dunlins  arrive  in  flocks  about  April 
and  May,  when  they  have  assumed  their  full  summer  dress  with 
black  breasts. 

Mr.  Chapman  found  a  nest  of  the  Dunlin  in  the  marisma 
containing  four  eggs. 

Bill  very  slightly  bent  down  towards  the  tip  and  longer  than  head  ;  bill, 
legs,  and  feet  black. 

Summer.  Belly  black;  general  colour  above  reddish  brown,  marked  with 
black. 

Winter.  Belly  white.  General  colour  above  ashy  grey.  Length 
8  inches. 

308.  Tringa  striata,  Linnaeus.    The  Purple  Sandpiper. 

This  species,  which  might  perhaps  be  well  termed  the  Rock- 
Sandpiper,  from  the  habit  of  frequenting  rocky  and  stony  coasts, 
is  altogether  omitted  by  Favier  as  a  Moorish  bird. 

They  are  not  common  in  autumn  and  spring. 

Bill  as  in  Dunlin,  but  base  dull  yellow. 

Rump  and  upper  tail-coverts  nearly  black,  with  a  purplish  tinge;  seventh 
to  ninth  secondaries  nearly  white  ;  legs  rather  short  and  dull  yellow. 
Length  8  inches. 

Young  have  feathers  above  with  white  margins. 

309.  Tringa  mirmta,  Leisler.    The  Little  Stint. 

This  small  Sandpiper  is  not  mentioned  by  Favier  as  occurring 
on  the  Moorish  coast ;  but  is  found  there  from  autumn  to  spring, 
and  we  fell  in  with  vast  flocks  at  Meshree  el  Haddar  on  the  26th 
of  April  in  company  with  Dunlins  and  Ringed  Plovers  ;  they  had 
then  attained  their  full  breeding-dress.  Mr.  Irby,  60th  Rifles, 
obtained  them  at  Tangier  on  the  18th  of  March,  1894. 


MACHETES  PUGNAX.  283 

On  the  Spanish  side,  the  Little  Stint  occurs  in  like  manner ; 
but  I  never  saw  any  large  numbers  near  Gibraltar. 

Plumage  above  much  as  in  Dunlin  both  in  summer  and  winter,  but  bill 
quite  straight. 

Outer  tail-feathers  ashy  brown  ;  legs  and  feet  black.     Length  6  inches. 

310.  Tringa  temmincki,  Leisler.    Temminck's  Stint. 

This  Stint  is  not  referred  to  by  Favier  or  recorded  by  Mr. 
Drake  as  occurring  on  the  African  side  of  the  Straits,  where, 
however,  they  are  of  course  to  be  found  as  on  the  Spanish  side, 
being  common  there  during  the  winter  and  found  in  small  parties 
of  from  six  to  a  dozen  or  more  in  number.  They  keep  much  to 
the  muddy  banks  of  tidal  rivers,  especially  frequenting  the 
salinas,  or  salt-pits ;  a  sure  locality  for  them  used  to  be  the 
abandoned  or  unused  salinas  near  Palmones,  between  Algeciraz 
and  Gibraltar,  and  they  are  very  seldom  seen  alone,  being  usually 
tame  and  easy  to  obtain.  I  failed  to  observe  any  later  than  the 
month  of  March  ;  but  no  doubt  they  further  prolong  their  stay  in 
southern  parts. 

Bill  straight ;  legs  and  feet  greenish  brown. 
Outer  tail-feathers  pure  white.     Length  of  inches. 

311.  Machetes   pugnax   (Linnaeus).     The   Ruff   (male).     The 

Reeve  (female). 

Moorish.  Habeeb  el  tchibeeb  (the  friend  of  the  Godwit). 

"  This  species  is  only  observed  near  Tangier  when  on  migration, 
crossing  to  Europe  during  March,  returning  in  July,  August,  and 
September.  Those  which  return  in  the  last  days  of  July  still 
exhibit  traces  of  the  breeding-plumage." — Favier. 

The  greater  number  of  Ruffs  pass  northwards  through  Anda- 
lucia  in  April;  but  flocks  occasionally  occur  during  January, 
February,  and  March,  some  passing  as  late  as  the  last  week  in 
May.  The  males,  or  "  Ruffs,"  are  then  in  their  inconvenient- 
looking  nuptial  plumage,  but  they  have  not  yet  been  detected 
nesting  so  far  south. 


284  SCOLOPACID.E. 

Axillaries  white,  no  white  on  primaries,  secondaries,  or  central  upper 
tail-coverts  ;  outside  tail-feathers  mottled  ;  hill  straight,  as  long  as  head. 

Male  in  spring.  With  a  ruff  of  various  shades  from  black  to  pure  white 
and  chestnut. 

Female,  and  male  in  winter.  Without  ruff.  Legs  yellowish  brown. 
Length  :  male  12^  inches,  female 


312.  Totanus  ochropus  (Linnaeus).    The  Green  Sandpiper. 

Spanish.  Lavandera. 

"  This  species  is  not  uncommon  in  winter  around  Tangier, 
frequenting  the  edges  of  lakes  and  the  banks  of  rivers  alone  or 
in  pairs.  They  depart  northwards  during  February  and  March, 
reappearing  by  August  and  September."  —  Favier. 

The  Green  Sandpiper,  a  bird  in  which  I  have  from  my  boyish 
days  in  Norfolk  always  taken  a  special  interest,  is  in  Anda- 
lucia,  as  in  England,  extremely  irregular  and  uncertain  in  its 
movements,  changing  its  ground  continually,  and  fluctuating 
greatly  in  numbers  ;  days  elapse  without  seeing  a  single  bird, 
and  suddenly  several  appear,  but  are  seldom  observed  in  any 
greater  number  than  two  or  three  together  ;  generally  they  are 
solitary  in  their  habits,  and  without  exception  frequent  shores  of 
freshwater  lakes,  ponds,  and  streams.  The  loud  note  of  this 
Sandpiper  and  the  white  tail-coverts,  which  show  markedly  on 
the  wing,  can  hardly  fail  to  cause  recognition.  The  curious 
fact  of  their  nidification  on  trees  in  old  nests  of  other  birds 
has  probably  led  to  their  nests  in  many  countries  being  over- 
looked ;  and  who  can  tell  that  they  may  not  yet  be  found 
breeding  in  Andalucia  1 

The  Green  Sandpiper  is  most  common  in  the  winter  months 
near  Gibraltar  ;  and  the  only  month  in  which  I  have  not  seen 
any  has  been  July  ;  but  then  I  had  no  opportunity  of  so  doing. 
This  Sandpiper  always  has  a  musky  odour. 

Bill  straight,  longer  than  head,  as  in  the  genus  Totanus. 

Above  greenish  brown,  with  minute  white  spots  ;  below  white  ;  rump 
and  outer  tail-feathers  on  each  side  white  ;  axillaries  brownish  black,  with 
narrow  white  bars  ;  legs  dark  green.  Length  9|  inches. 


TOTANUS  HYPOLEUC'US.  285 

313.  Totanus  glareola  (J.  F.  Gmelin).    The  Wood-Sandpiper, 
This  bird  is  not  noticed  by  Favier  as  occurring  in  Morocco  ; 

but  there  were  plenty  to  be  seen  towards  the  end  of  April  at  the 
lakes  of  Ras  el  Doura  and  other  swamps  in  that  country ;  and 
near  Gibraltar  I  have  observed  them  frequently  on  passage  from 
the  9th  of  March  to  the  beginning  of  the  month  of  May. 

Being,  as  far  as  my  observations  go,  entirely  a  freshwater 
Sandpiper,  the  species  most  resembles  in  habits  the  Marsh- 
Sandpiper  (T.  stagnatilis). 

Above  greenish  brown,  each  feather  margined  with  buffy-white  spots  ; 
upper  tail-coverts  white  ;  axillaries  white,  sometimes  with  a  few  dusky 
bars  ;  legs  pale  olive,  1^  inches  long.  Length  7£  inches. 

314.  Totanus  hypoleucus   (Linnaeus).     The  Common  Sand- 

piper. 

Spanish.  Lavandera  chica. 

"  This  is  the  most  common  of  the  Sandpipers  around  Tangier, 
passing  north  during  April  and  May.  They  are  seen  returning 
in  August,  September,  and  October." — Favier. 

The  Common  Sandpiper  in  Andalucia  prefers  the  banks  of 
running  streams  and  salt  or  tidal  marshes,  being  not  much 
noticed  in  freshwater  marshes  or  about  stagnant  pools.  Near 
Gibraltar,  particularly  when  on  passage  in  spring,  they  greatly 
frequent  the  sea-coast  wherever  there  is  much  seaweed  thrown 
up  by  the  tide ;  and  I  have  repeatedly  observed  them  on  rocks, 
like  the  Purple  Sandpiper. 

The  present  species  swarms  about  the  Straits  in  March  and 
April,  passing  in  lots  of  four  or  five  together.  I  have  no  record 
of  any  in  November,  but  saw  one  on  the  24th  of  October  and 
one  on  the  7th  of  December ;  they  are  not  abundant  in  spring 
until  the  month  of  March,  their  passage  being  at  its  height  about 
the  15th  of  April.  Though  I  never  succeeded  in  finding  a  nest, 
am  positive  some  remain  to  breed,  as  in  the  end  of  May, 
near  the  mouth  of  the  Guadiaro,  a  pair,  from  their  manner, 


286-  SCOLOPACID^. 

were  certainly  nesting ;  but  all  my  efforts  to  discover  the  situation 
were  futile.     Arevalo  says  they  breed  in  Andalucia. 

Above  greenish  brown  ;  axillaries  white  ;  secondaries  brown,  with  white 
bases  and  tips.  Length  7|  inches. 

315.  Totanus  cinereus  (Gtildenstadt).    The  Terek  Sandpiper. 

This  bird  has  been  once  obtained  near  Malaga.  Is  a  fresh- 
water Sandpiper,  and  an  Eastern  species,  ranging  to  in  winter  as 
far  south  as  S.  Africa  and  Australia. 

Bill  slightly  turned  up,  as  in  Greensbank;  secondaries  marked  with 
white  ;  axillaries  white.  Length  8£  inches. 

316.  Totanus  stagnating,  Bechstein.     The  Marsh-Sandpiper 

or  Lesser  Greenshank. 

This  eastern  species  has  been  obtained  near  Malaga  and 
Valencia  (Arevalo). 

Is  a  freshwater  Sandpiper,  and  a  miniature  Greenshank,  having  grey 
secondaries,  but  a  straight  bill.  Length  10  inches. 

317.  Totanus  oanesoens  (J.  F.  Gmelin).    The  Greenshank. 

All  that  Favier  has  to  say  of  this  bird  is  that  it  is  met  with  on 
passage,  "  returning  south  during  the  month  of  October  to  winter 
probably  in  the  interior  of  Africa " ;  and  as  the  Greenshank  is 
recorded  by  Andersson  as  common  in  Damara  Land,  Favier  was 
not  much  out  in  his  supposition.  He  does  not,  however,  mention 
the  date  of  its  vernal  migration,  which  takes  place  in  March, 
April,  and  May,  the  birds  being  most  frequently  seen  during  the 
latter  month,  and  the  latest  recorded  by  me  was  the  22nd  of 
May.  I  have  also  noticed  them  in  November  and  January,  but 
never  in  any  numbers ;  in  all  probability  their  chief  line  of 
migration  lies  further  to  the  east. 

The  Greenshank  is  a  very  noisy  bird,  and  sure  to  attract  notice 
by  its  loud  whistling  cry,  which,  as  is  well  known,  consists  of  two 
notes. 


TOTANUS  FUSCUS.  287 

Secondaries  uniform  grey  ;  bill  black,  longer  than  'head,  and  slightly 
turned  upwards  ;  legs  olive-green. 

In  breeding-season  has  blackish  marks  on  back  and  black  spots  on  flanks, 
most  developed  in  the  female.  Length  13-14  inches. 

318.  Totanus  calidris  (Linnaeus).    The  Common  Redshank. 
Andalucian.  Archibebe. 

Favier  states  that  "  this  Redshank  is  very  abundant  near 
Tangier,  in  small  lots,  which  frequent  the  edges  of  rivers  and 
lakes,  and  mostly  pass  northwards  during  March  and  April, 
returning  to  remain  for  the  winter  in  September  and  October. 
Some,  however,  remain  in  the  country  for  the  breeding-season.'' 

We  found  the  Common  Redshank  in  some  numbers  at  the 
lakes  of  Ras  el  Doura  towards  the  end  of  April ;  and  they  were 
then  evidently  beginning  to  nest,  but  were  not  in  anything  like 
the  quantity  which  breed  in  some  parts  of  the  marisma  of  the 
Guadalquivir,  where  they  breed  a  little  later  than  the  Peewit, 
which  is  there  the  earliest  marsh-nesting  bird.  In  Andalucia 
this  Redshank  is,  though  frequently  seen  in  winter,  and  much 
too  often  for  the  sportsman,  chiefly  migratory,  passing  north  in 
great  abundance  mostly  towards  the  middle  of  April,  when  many 
are  to  be  seen  and  heard  shrieking  out  their  three  notes  about 
the  old  salinas  or  abandoned  salt-pits  at  Palmones,  near  Gibraltar ; 
and  a  great  many  fall  victims,  to  appear  ultimately  in  the  market ; 
but  they  are  quite  unfit  to  eat. 

Secondaries,  axillaries,  and  upper  tail-coverts  white;  legs  and  feet  red  in 
adults,  yellowish  in  young.  Length  10-11  inches. 

319.  Totanus  fuscus  (Linnaeus).    The  Dusky  Redshank. 

Favier's  notes  relative  to  this  species  are  as  follows : — "  Fre- 
quents the  vicinity  of  salt  marshes  near  Tangier  during  the 
months  of  September  and  October " ;  but  the  brevity  of  his 
remarks  on  most  of  the  Waders  and  aquatic  birds  would  lead 
one  to  suspect  that  Favier,  like  many  other  Frenchmen  and  all 
Spaniards,  had  a  cat-like  antipathy  to  water. 


288 

On  the  Spanish  side  of  the  Straits  I  never  shot  a  specimen  of 
the  Dusky  Redshank ;  but  it  occasionally  occurs  in  spring  and 
autumn.  Their  whistle  is  a  single  note. 

Larger  than  last ;  secondaries  white,  barred  with  grey  ;  upper  tail-coverts 
white,  barred  with  blackish  brown  ;  legs  red.  Length  12  inches. 

320.  Limosa  lapponica  (Linnseus).    The  Bar-tailed  Godwit. 

Moorish.  Tchibeeb  (Favier). 

"  This  species  is,  during  passage,  nearly  as  numerous  in  the 
vicinity  of  Tangier  as  L.  cegocephala.  They  arrive  from  the 
north  during  September,  and,  passing  on  further  south,  return 
and  cross  over  to  Europe  during  the  months  of  February,  March, 
and  April.  The  chasseurs  of  Larache  call  this  Godwit  Bou- 
meraisa  indiscriminately  with  the  other  species." — Favier. 

The  Bar-tailed  Godwit,  chiefly  from  frequenting  salt-marshes 
and  estuaries  of  rivers,  is  not  noticed  near  Gibraltar  in  such 
numbers  as  the  larger  Black-tailed  Godwit,  and  is  only  observed 
on  passage.  The  latest  date  I  have  of  seeing  them  in  spring 
was  on  the  10th  of  May ;  and  the  earliest  date  of  their  appearance 
recorded  in  autumn  was  the  21st  of  September;  but  no  doubt 
they  pass  much  sooner  than  this.  I  did  not  observe  any  during 
the  winter  months. 

Bill  as  in  next  species  very  long. 

Rump  nearly  white  ;  axillaries  barred  with  black  and  white  ;  tail  barred 
in  spring  and  in  young  birds  ;  adults  in  winter  have  tail  ash-grey,  marbled 
at  the  base,  long  tail-coverts  barred. 

Female  Godwits  much  larger  than  males. 

In  summer.  Below  pale  red,  darker  in  male. 

In  winter.  Below  white,  above  greyish  brown  ;  feathers  with  dark  shaft- 
streaks.  Length  15-16  inches. 

321.  Limosa    belgica    (J.  F.  Gmelin).      Tlie    Black  -  tailed 

Godwit. 

Moorish.  Tchibeeb  (Favier).     Andalucian.  Abujeta,  Sarseruelo. 

"  This  Godwit  is  found  on  passage  near  Tangier  in  abundant 

flocks,  migrating  to  the  north  during  the  months  of  February  and 


LIMOSA  BELGICA.  289 

March  ;  they  are  observed  returning  in  August  and  September." 
— Favier. 

Favier  also  further  asserts  that  this  species  occasionally  remains 
to  breed  in  Morocco — a  statement  which  sounds  improbable,  and 
therefore  to  be  received  with  reserve,  though  it  would  not  perhaps 
be  more  surprising  than  the  fact  of  the  Crane,  Peewit,  and  the 
Dunlin  nesting  so  far  south  as  they  have  been  proved  to  do. 

The  Black-tailed  Godwit  appears  in  Andalucia  during  February, 
in  bands  of  from  four  or  five  to  as  many  as  two  or  three  hundred, 
frequenting  the  grassy  marshes  or  rather  inundated  ground  about 
Casas  Viejas  and  the  marisma.  They  are  very  restless,  and 
continually  on  the  move,  uttering  their  loud  cries.  As  they  are 
usually  rather  wary  and  difficult  to  get  a  shot  at,  the  best  chance 
of  obtaining  any  is  either  to  lie  up  for  them,  or,  in  Spanish 
fashion,  to  use  a  stalking-horse.  Their  numbers  vary  considerably 
in  different  years ;  and  they  do  not  seem  to  stay  long  in  the  same 
district,  as  some  hundreds  may  be  noticed  one  day  and  hardly 
any  on  the  next.  The  passage  continues  far  into  the  month  of 
March,  by  which  time  they  are  well  advanced  in  their  rufous 
breeding-plumage  ;  and  this  ruddy  appearance  shows  much  when 
they  are  on  the  wing.  The  earliest  assumption  of  this  dress  was 
observed  on  the  24th  of  February.  Immense  quantities  are 
brought  into  Seville  market  for  sale  during  March ;  and  the 
latest  seen  there  was  on  the  6th  of  April.  In  that  district  their 
local  name  is  Sarseruelo  ;  but  about  Casas  Viejas  they  are  known 
as  Abujeta,  evidently  a  word  of  Moorish  origin. 

They  are  excellent  eating  and  fully  deserve  the  derivation 
which  has  been  given  of  their  English  name  "  Godwit,"  i.  e. 
good  food  (from  "  good,"  and  "  wihta,"  animal). 

I  have  no  note  of  their  autumnal  migration  ;  but  occasionally 
they  are  met  with  in  winter,  usually  solitary  birds,  as  I  killed 
one  on  the  5th  of  December  at  Tapatanilla,  and  have  seen  others 
now  and  then  obtained  in  January. 

Larger  than  last ;  upper  tail-coverts  white  ;  tail  black,  with  white  base  ; 

U 


290  SCOLOPACID.E. 

white  bar  across  wing  ;  axillaries  white,  but  sometimes  barred  with 
brown. 

In  summer.  Breast  reddish,  belly  white,  barred  with  blackish  brown. 

In  winter.  Ashy  brown  above,  beloAv  greyish  white.  Length  16-20 
inches. 


322.  Numenius  arquata  (Linnaeus).    The  Common  Curlew. 

Moorish.  Bou-khalal  (Father  of  the  toothpick  or  pin).  Spanish. 
Zarapito. 

"This  bird  is,  near  Tangier,  only  a  winter  resident,  which 
arrives  in  September  and  October  and  leaves  during  March. 
They  frequent  the  mouths  of  rivers  and  the  sea-shore  in  large- 
numbers,  but  they  are  very  wild  and  difficult  to  get  a  shot  at." 
— Favier. 

When  at  Larache  towards  the  end  of  April  I  observed  several 
Common  Curlews  ;  and  a  Spaniard  who  resided  there  asserted 
that  they  nested  near  the  town.  No  doubt  if  any  Curlew  breeds 
there  it  would  be  the  next  small  species. 

The  Common  Curlew  is  very  plentiful  near  Gibraltar  during  the 
winter  months,  being,  perhaps,  most  frequent  in  February ;  but  is 
very  wary,  as  everywhere  else  in  the  world  that  I  have  met  with  it. 

Bill  as  in  next  three  species,  very  long  and  curved  downwards. 
Crown  pale  brown,  regularly  streaked  with  dark  brown;  axillaries  white, 
barred  with  dark  brown;  rump  paler  than  back.     Length  21-26  inches. 

323.  Numenius   tenuirostris,  Vieillot.     The  Mediterranean 

Curlew. 

This  small  Curlew  is  said  to  occur  in  spring  and  autumn  about 
the  Straits.  I  never  met  with  any,  but  Arevalo  records  them 
from  the  Province  of  Malaga. 

Capt.  Savile  Reid  mentions  ('Ibis,'  1885)  that  this  bird  was 
abundant  during  the  winter  of  1884-85  in  the  Laraish  Valley,  in 
flocks  of  from  twenty  to  a  hundred.  He  obtained  specimens 
there,  and  also  saw  numerous  flights  at  Meshree  el  Haddar ;  he 


LAKID.E.  291 

further  describes  the  note  as  resembling  that  of  the  Common 
Curlew,  "  but  not  so  loud  or  musical." 

Crown  uniformly  striated,  no  pale  streak ;  axillaries  white ;  flanks  marked 
with  conspicuous  dark  pear-shaped  spots.  About  size  of  the  Whimbrel. 

324.  Numenius  phseopus  (Linnaeus) .    The  "Whimbrel. 
Favier's  notes  on  this  species  are  the  same  as  on  the  Curlew, 

except  that  he  adds  "  this  bird  arrives  earlier  from  the  north,  and 
though  very  common,  does  not  remain  in  the  neighbourhood  of 
Tangier  for  the  winter,  but  passes  on  further  south." 

On  the  Spanish  side  of  the  Straits,  the  Whimbrel,  plentiful  in 
autumn  and  spring  up  to  the  end  of  April,  is  occasionally  seen 
in  winter;  and,  as  elsewhere,  is  far  less  wary  and  difficult  to 
approach  than  the  Curlew. 

Crown  dark  brown,  with  pale  streak  along  the  top ;  in  some  young  birds 
this  streak  is  almost  obsolete,  and  the  brown  on  head  has  a  glossy  sheen ; 
axillaries  white,  barred  more  or  less  with  brown.  Length  16-18  inches. 

325.  Numenius  hudsonicus,  Latham.    The  American  Whim- 
brel. 

A  specimen  of  this  Whimbrel  was  obtained  by  Lord  Lilford  in 
the  Coto  de  Donana  on  the  3rd  of  May,  1872  ('  Ibis,'  1873, 
p.  98). 

About  size  of  the  last.  Crown  pale  brown,  with  pale  streak  along  the 
centre ;  axillaries  rufous. 

Order   GAVIJ3.  Family  LARIDJE. 

Subfamily  STERNIN^E.     Legs  and  feet  small  and  weak ; 
bill  as  long  as  or  longer  than  head. 

326.  Hydrochelidon  nigra  (Linnaeus).    The  Black  Tern. 
Spanish.  Cencerillo,  Paino. 

"  This  Tern  is  abundant  near  Tangier  when  on  passage, 
crossing  the  Straits  in  large  nights  during  May,  and  returning  in 

TT  O 

U   - 


292  LARID.E. 

September    and   October.      They   are   not   seen   in    the  winter 
months." — Favier. 

The  Black  Tern,  on  the  Spanish  side,  begins  to  appear  about 
the  end  of  April.  Is  rather  later  than  the  Whiskered  Tern, 
nesting  also  later  in  the  same  situations  as  that  species,  but  not 
in  such  numbers.  I  have  noticed  quantities  crossing  the  Straits 
on  the  10th  of  May,  and  they  are  often  seen  on  arrival,  hawking 
about  over  cornfields  and  low  ground  near  water. 

Head,  neck,  breast,  and  belly  black  ;  vent  and  under  tail-coverts  white  ; 
under  wing-coverts  pale  grey  ;  rest  of  plumage  slate-grey  ;  bill  black;  legs 
reddish  brown.  Length  10  inches. 

Autumn.  Forehead,  throat,  and  nape  white  ;   below  barred  with  white. 

327.  Hydrochelidon  leucoptera  (Schinz).    The  White-winged 

Black  Tern. 

On  the  African  side  the  only  instance  recorded  of  the  occur- 
rence of  this  bird  is  a  single  specimen  shot  in  May,  1869,  at  Sharf 
el  Akab,  near  Tangier.  It  is  not  included  in  Favier's  list ;  and 
we  never  met  with  any  on  either  side  of  the  Straits,  but  three 
shot  in  the  marisma  are  in  Lord  Lilford's  collection  ;  probably 
they  rarely  wander  so  far  west.  I  have  several  times  imagined 
that  I  saw  some,  but  upon  shooting  the  birds  found  them  to  be 
fine  old  H.  hybrida,  which,  when  flying  in  the  bright  sunshine 
and  showing  the  blackish  belly,  is  very  apt  to  be  mistaken  for 
the  present  species. 

Head,  neck,  underparts,  and  under  wing-coverts  black ;  vent,  tail,  and 
tail-coverts  white ;  bill  dark  red  ;  legs  scarlet.  Length  9^  inches. 

328.  Hydrochelidon  hybrida  (Pallas).    The  Whiskered  Tern. 
Spanish.  Paino  mayor  (Malaga)  (Arevalo). 

11  This  Tern  is  scarce  near  Tangier,  and  seen  only  on  passage 
during  April,  returning  south  in  August.  Immense  numbers  are 
found  breeding  at  the  lakes  of  Eas  el  Doura,  where,  nesting 
together  in  vast  colonies,  they  bear  the  local  name  of  Mershifc." 
— Favier. 


UJ 

2 

O 

I 


CO 

z 

K 
UJ 

J- 
o 

UJ 

o: 

UJ 

^ 

CO 

I 


STERNA  ANGLICA.  293 

On  the  Spanish  side  the  Whiskered  Tern  arrives  about  the 
middle  of  April,  and  is  only  seen  near  Gibraltar  on  migration. 
Hovering  over  every  swamp  and  wet  spot,  they  soon  pass  on 
to  their  breeding-haunts  in  the  Laguna  de  la  Janda  and  the 
marisma,  where,  among  rushes  and  sedges,  they  nest  about  the 
middle  of  May  in  colonies.  Like  other  Terns,  they  lay  three 
eggs  ;  of  a  pale  green  ground-colour,  with  variable  markings  of 
blackish  brown. 

I  saw  a  large  flock  of  these  Terns  on  the  sea  near  Cadiz  on 
the  18th  of  July.  Verner  visited  a  breeding-colony  of  these 
birds  at  La  Janda  on  the  7th  of  May,  1875,  and  found  several 
hundred  nests  floating  on  the  top  of  the  water  ;  they  were  simple 
platforms  of  reeds  and  rushes,  and  were  kept  from  drifting  to 
some  extent  by  the  young  rushes  growing  up  in  the  water.  Only 
two  nests  contained  a  single  egg.  Five  days  later  over  thirty 
nests  had  eggs.  In  the  interval  between  the  visits  a  strong  wind 
had  arisen  and  blown  many  of  the  Terns'  nests  along  the  water 
till  they  were  packed  in  a  dense  mass  on  the  lee  side  of  the 
laguna. 

Forehead,  crown,  and  nape  black  ;  chin  and  sides  of  neck  white  ;  belly 
very  dark  brown,  almost  black;  rest  of  plumage  grey  or  white;  under  wing- 
coverts  white;  bill  dark  red;  legs  vermilion.  Length  11^  inches. 


329.  Sterna  ang-lica,  Montagu.    The  Gull-billed  Tern. 

Spanish.  Cagalo  (Arevalo). 

Strangely  enough,  no  mention  is  made  of  this  Tern  in  Favier's 
MS.;  but  we  found  them  in  great  numbers  about  the  lakes  of 
Ras  el  Doura  towards  the  end  of  April.  As  far  as  we  could 
ascertain  from  the  Arabs,  they  said  that  these  birds  remained  in 
the  neighbourhood  and  bred  a  little  later  on  in  the  season. 
Essentially  a  freshwater  or  marsh-frequenting  species,  I  never 
noticed  the  Gull-billed  Tern  on  the  sea-coast.  Some  of  those 
we  shot  had  been  feeding  on  green  frogs  ;  their  note,  loud  and 
frequently  repeated,  is  (as  near  as  I  can  render  it)  Jcuh-wuJc, 
kuh-wuk.  I  never  noticed  the  present  species  about  Gibraltar, 


294  LARID.E. 

but  it  occurs  in  the  marshes  of  the  Guadalquivir  towards  San 
Lucar,  breeding  there  in  considerable  numbers,  and  its  eggs  from 
thence  have  been  supposed  to  belong  to  Larus  melanocephalus. 

Head  and  bill  black;  lower  mandible  slightly  angnlated  or  "Gull-billed"; 
tail  bluish  grey.  Length  13-14  inches. 

330.  Sterna  caspia,  Pallas.    The  Caspian  Tern. 
Spanish  for  all  Terns.  Golondrina  de  mar. 

This  large  Tern  is  stated  by  Favier  to  be  "  very  rare  near 
Tangier,"  he  having  only  obtained  a  single  specimen  (in  February, 
1844),  which  occurrence  I  can  supplement  by  one  which  occurred 
in  the  winter  of  1869. 

Near  Gibraltar  I  did  not  meet  with  any,  though  two  were 
obtained  by  Lord  Lilford  near  the  Isla  Menor,  below  Seville,  in 
April,  1883.  They  are  only  accidental  stragglers  so  far  west,  but 
further  east  in  Spain,  near  Mar  Menor,  they  are  reported  by 
Guirao  to  be  common. 

The  largest  of  the  Terns. 

Head  black  ;  tail  white  ;  bill  coral-red,  dark  at  tip;  legs  black.  Length 
19-21  inches. 

331.  Sterna  maxima,  Boddaert.    The  Royal  Tern. 

This  Sea-Tern  was  once  obtained  by  Favier  in  the  Straits  and 
is  described  in  his  MS.  The  specimen,  which  I  purchased  from 
his  successor,  was  an  adult  bird  in  winter  plumage,  and  is  now  in 
the  possession  of  Lord  Lilford.  Two  specimens  shot  near  Tangier 
in  December,  1882,  are  in  Mr.  J.  J.  Dalgleish's  collection,  to 
wbom  I  am  obliged  for  this  information. 

Crown  black  and  slightly  crested  ;  bill  orange-red.     Length  19  inches. 

332.  Sterna  cantiaca,  J.  F.  Gmelin.    The  Sandwich  Tern. 

Spanish.  Charran  (Malaga)  (Arevalo). 

"  This  Tern  is  seen  near  Tangier  in  abundant  flocks  from 
November  to  February." — Favier. 


STERXA  MACKURA.  295 

They  were  very  numerous  at  the  mouth  of  the  river  at  Larache 
during  April. 

The  Sandwich  Tern  is  very  common  in  the  Straits  in  autumn, 
winter,  and  spring.  Sometimes  thirty  or  forty  may  be  noticed 
sitting  together  on  the  small  isolated  rocks  near  Cabrita  Point, 
and  will  allow  a  boat  to  approach  within  a  few  yards.  They  pass 
north  about  the  first  week  in  April,  when  I  killed  an  old  male 
tinted  on  the  breast  and  under  wing-coverts  with  a  beautiful 
pink  blush,  just  as  is  sometimes  found  in  the  spring  on  old  males 
of  the  Brown-headed  Gull  (Lanes  ridilundus).  They  nest  in 
the  marisma  and,  according  to  Arevalo,  near  Malaga. 

Head  black  ;  tail  white  ;  bill  black,  with  yellow  tip  ;  legs  black.  Length 
15  inches. 

333.  Sterna  media,  Horsfield.    The  Allied  Tern. 

"  This  species  is  one  of  the  least  common  of  the  Terns  near 
Tangier,  and  only  occasionally  met  with.  Further  south,  in  the 
vicinity  of  Larache,  it  is  more  frequently  seen ;  and  I  found  it 
there  during  September,  October,  and  November,  in  company 
with  S.  cantiaca,  which  species  it  resembles  in  habits." — Favier. 

This  Tern  occurs  in  the  Straits  in  spring.  Two,  both  males, 
were  shot  near  Tarifa  on  the  20th  of  April,  1873,  and  many  have 
been  obtained  near  Tangier.  Probably  they  breed  on  the  coast. 

This  bird  is  very  much  like  the  Sandwich  Tern  (S.  cantiaca),  but  is  a 
trifle  larger  and  has  the  bill  yellow.  I  found,  on  comparing  male  specimens 
shot  on  the  same  day,  that  it  differs  from  that  species  also  in  having  the  bill 
stouter  in  proportion,  and  the  lower  mandible  slightly  angulated,  or  "  Gull- 
billed."  The  feathers  of  the  black  crest  are  more  elongated,  and  the  upper 
tail-coverts  and  tail  are  grey,  the  same  colour  as  the  back.  The  primaries 
underneath  are  more  broadly  marked  with  grey  next  the  shafts  ;  and  the 
tarsus  is  rather  longer. 

334.  Sterna  macrura,  Naumaun.    The  Arctic  Tern. 

I  obtained  this  Tern  in  winter  plumage  in  the  Straits  of 
Gibraltar;  and  there  is  no  doubt  that  they  occur  regularly  on 


296  LAKID^E. 

migration ;  but  it  was  very  difficult  to  get  specimens  of  the  Terns 
for  identification. 

Crown  black  ;  below  as  grey  as  the  back.  First  primary  with  the  grey 
stripe  along  the  shaft  of  the  inner  web  no  broader  than  the  outer  web  ;  bill 
and  legs  coral-red;  legs  shorter  than  in  next.  Length  15  inches. 

335.  Sterna  fluviatilis,  Nauuiann.    The  Common  Tern. 

Favier  includes  this  species  in  his  notes,  and  states  that  they 
are  found  near  Tangier  in  large  flights  on  the  coast  during 
migration,  passing  south  during  September  and  October.  Pos- 
sibly he  may  mean  the  previous  species. 

The  Common  Tern  is  frequently  seen  in  autumn  and  spring  in 
the  Straits,  and  may  very  possibly  be  found  nesting  near  Cadiz. 

Crown  black,  below  white. 

First  primary  with  the  grey  stripe  along  the  shaft  of  the  inner  web  twice 
as  broad  as  the  outer  web  ;  bill  and  legs  coral-red  ;  bill  blackish  at  tip. 
Length  14  inches. 

336.  Sterna  minuta,  Linnaeus.    The  Little  Tern. 

Spanish.  Catalinita  (Arevalo). 

"This  small  Tern  is  seen  near  Tangier,  passing  in  small 
flights  along  the  coast  and  on  the  rivers  and  lakes.  They  arrive 
during  May,  and  return  in  September,  some,  however,  remaining 
in  the  country  to  breed.  They  all  retire  south  for  the  winter." — 
Favier. 

The  Little  Tern  occurs  on  the  Spanish  as  on  the  Moorish  side 
and,  keeping  to  the  sea-coast,  is  the  latest  to  arrive  of  all  the 
family.  They  are  nowhere  very  abundant ;  but  a  few  nest  near 
the  mouth  of  the  Guadiaro  about  the  end  of  May,  as  well  as  in 
other  localities  on  the  coast. 

The  earliest  date  on  which  noticed  was  on  the  10th  of  May ; 
and  the  latest  was  on  the  25th  of  October. 

Primaries  white,  with  black  tips,  edged  on  the  inner  web  with  black  and 
usually  on  the  outer  web.  Length  15-16  inches. 


LARUS  MELANOCEPIIALU3.  297 

Subfamily  L ARIN^.     Legs  and  feet  large  and  strong ;  bill 
shorter  than  head. 

Moorish  for  Gulls.  Garvia  or  Kerzeit.     Spanish.  Gaviota. 

337.  Larus  minutus,  Pallas.    The  Little  Gull. 

Favier  only  mentions  having  once  obtained  this  diminutive 
Gull  near  Tangier,  in  February,  1854.  I  have  seen  them  in 
some  numbers,  though  very  irregularly,  in  winter.  Verner  says 
a  small  flock  frequented  the  Laguna  de  las  Jabas  in  February, 
1892.  I  saw  two  sitting  on  the  bowsprit  of  a  yacht  anchored  at 
Gibraltar  on  the  24th  of  March,  1894. 

Spring.  Head  black. 

Winter.  Head  white,  with  a  few  dusky  spots. 

Young.  Head  brownish,  marked  with  black  and  grey. 

Smallest  Gull,  length  offing  under  9^  inches.     Length  10-11  inches. 

338.  Larus  ridibundus,  Linnaeus.    The  Brown-headed  Gull. 
According  to  Favier  this   Gull  is  the  most  common  species 

around  Tangier,  arriving  chiefly  during  November,  and  departing 
north  in  March. 

Is  found  as  above  on  the  Spanish  side,  and  was  noticed  as  late 
as  the  10th  of  April,  1894,  at  the  Laguna  de  la  Janda. 

Head  sooty  brown  from  about  March  1st  to  August  1st. 

Winter.  Head  white,  with  grey  patch  behind  eye. 

Young.  Marked  above  with  brown,  and,  as  in  all  Hack-headed  Gulls,  has 
a  blackish  bar  at  the  end  of  tail. 

Primaries  white,  with  black  tips,  edged  on  the  inner  web  with  black  and 
usually  on  the  outer  web.  Length  15-16  inches. 

339.  Larus  melanocephalus,  Natterer.    The  Adriatic  Black- 

headed  Gull. 

This  Gull  may  occasionally  occur  in  the  Straits  in  winter ;  but 
I  never  obtained  one,  and  Lord  Lilford  tells  me  that  he  never  saw 
any  westward  of  Malaga. 

Mr.  H.  Saunders  writes  to  me  as  follows  : — "  There  has  been  a 


298  LARID^E. 

good  deal  of  confusion  about  the  supposed  breeding  of  this  bird 
in  Spain,  and  as  I  have  in  some  degree  contributed  to  this  I  will 
do  my  best  to  clear  matters  up.  I  saw  and  have  specimens  of 
this  Gull  from  Malaga  and  also  from  the  Trocadero,  near  Cadiz, 
and  I  had  eggs — said  to  have  been  taken  with  the  birds  (which  I 
examined)  in  the  marisma.  I  also  saw  a  colony,  apparently 
breeding,  below  Huelva,  but  the  steamer  in  which  I  had  taken 
my  passage  whistled  me  back  before  I  could  land.  The  eggs  above 
mentioned  have  proved  to  be  those  of  the  Gull-billed  Tern ;  so 
were  those  obtained  by  Mr.  Abel  Chapman  ('Ibis,'  1884,  p.  86), 
who  shot  an  Adriatic  Gull  near  them  *  ;  so  were  those  acquired 
by  Mr.  Dresser  (save  one,  which  is  a  Sandwich  Tern's) ;  and  I 
do  not  know  of  an  authenticated  egg  of  L.  melanoceplialus  from 
Spain. 

"  Of  course  I  had  not  seen  the  eggs  when  I  inserted  the 
'  advanced  notes '  furnished  by  Mr.  Dresser  in  my  '  Manual  of 
British  Birds ' ;  the  whole  account  will  be  cancelled  in  the  next 
edition. 

"  I  think  it  is  probable  that  the  species  will  be  found  nesting 
along  the  coast  or  on  the  etangs  of  the  Landes  between  Bayonne 
and  Bordeaux,  because  the  birds  visit  St.  Jean  de  Luz  in  spring 
and  also  the  mouth  of  the  Gironde." 

Bill  stouter  than  last. 

In  summer.  Head  jet-black. 

Adults.  Primaries  white,  except  the  first,  which  has  the  outer  web 
margined  with  black. 

Young.  Outer  webs  and  shafts  of  first  five  primaries  dark  brown,  outside 
of  inner  web  of  third  primary  white.  Length  15  inches. 

340.  Larus  canus,  Linnaeus.    The  Common  Gull. 
This  Gull  is  not  mentioned  by  Favier,  but  is  during  some 
winters  common  in  the  Straits  of  Gibraltar. 

In  spring.  Head  and  neck  white  ;  mantle  pale  grey  ;  bill  and  legs 
greenish  yellow  ;  former  yellow  at  point. 

*  But  this  bird  was  not  brought  home  for  identification. 


LARUS  GELASTES.  299 

In  winter.  Head  and  neck  streaked  and  spotted  with  brown  ;  legs  pale 
brown. 

Young.  Head,  neck,  and  underparts  white,  mottled  with  light  brown  ; 
tail  with  black  band  at  end.  Primaries  black  or  dark  brown,  with  white 
spots  at  the  end  or  next  the  end.  Length  18  inches. 


341.  Larus  audouini,  Payraudeau.    Audouin's  Gull. 

This  Mediterranean  Gull  is  recorded  by  Natterer  as  having 
been  once  obtained  near  Tarifa,  and  has  since  then  occurred  at 
Malaga. 

Lord  Lilford,  who  discovered  them  breeding  in  1874  off  the 
south  of  Sardinia  on  a  rocky  island,  and  also  found  them  on  the 
island  of  Alboran  in  April,  1879,  kindly  gave  me  the  following 
notes  of  the  colours  of  the  soft  parts,  which  will  serve  to  distin- 
guish the  adults  of  this  species  : — 

"  Feet  and  legs  very  dark  grey,  claws  black.  Bill  brilliant 
coral-red,  with  one  broad  black  band.  Iris  brilliant  hazel,  pupil 
black.  Inside  of  mouth  pale  flesh-colour.  Eyelids  coral-red." 

The  following  are  the  measurements  (in  inches)  of  a  pair, 

taken  from  the  dried  skins  : — 

d.  $. 

Bill  from  gape 3'00  2'80 

Wing,  carpus  to  tip    15'75  15'75 

Tarsus  2'45  2'30 

Tail   6-00  6-00 

Length  20  inches. 

342.  Larus  gelastes,  Lichtenstein.    The  Slender-billed  Gull. 

Favier  only  records  a  single  specimen  of  this  Gull  as  obtained 
by  him  near  Tangier,  in  1852  ;  but  he  remarks  that  the  feet  of 
the  immature  birds  are  orange-yellow,  which  would  lead  to  the 
supposition  that  he  observed  them  more  often. 

They  are  not  uncommon  in  spring  on  the  coast  of  Southern 
Spain,  nesting  near  the  mouth  of  the  Guadalquivir,  where 


300  LARID.E. 

Verner,  when  with  the  Crown  Prince  of  Austria,  found  eggs 
about  the  27th  of  May,  1879. 

Bill  slender,  as  long  as  head. 

Head,  nape,  and  neck  white.  Bill,  eyelids,  legs,  and  feet  vermilion. 
Irides  white.  Length  16  inches. 

343.  Larus  argentatus,  J.  F.  Gmelin.    The  Herring-Gull. 
The  Herring-Gull  is  stated  by  Favier  to  be  "  as  common  near 

Tangier  during  winter  as  L.  ridibundus,  arriving  in  August, 
September,  and  October,  and  returning  north  in  March,  April, 
and  May." 

This  bird  and  the  Lesser  Black-backed  Gull  feed  in  large 
numbers  on  the  refuse  from  the  slaughterhouses  at  Gibraltar; 
and  it  is  not  uncommon  to  see  three  or  four  hundred  of  them 
together  there. 

Perhaps  some  of  the  immature  birds  remain  during  the 
summer ;  but  all  the  adults  disappear  by  about  the  15th  of 
April. 

Head  and  neck  white  ;  mantle  pale  grey  ;  legs  and  feet  flesh-colour  ; 
eyelids  yellow. 

Young.  Feathers  above  brown,  with  pale  edges  ;  tail  barred  with  brown  ; 
underparts  white,  streaked  with  brown.  Length  22-24  inches. 

344.  Larus  cachinnans,  Pallas.     The  Southern  or  Yellow- 

legged  Herring-Gull. 

Favier  remarks  that  "  this  Gull  is  not  very  common  near 
Tangier,  where  it  consorts  with  L.  fuscus  and  L.  argentatus" 

This  race  of  Herring-Gull  is  common  about  the  Straits  and 
southwards  from  Santander,  where  it  occurred  abundantly  in  May, 
1876. 

The  adult  birds  are  distinguishable  by  the  colour  of  the  legs,  which  are 
yellow,  as  in  L.  fuscus  ;  the  eyelids  are  scarlet,  the  back  being  much  lighter 
in  colour,  but  darker  than  in  the  next. 

The  size  of  the  three  species  is  about  equal,  and  the  birds  in  immature 
plumoge  appear  identical. 


RISSA  TRIDACTYLA.  301 

345.  Larus  fuscus,  Linnseus.    The  Lesser  Black-backed  Gull. 
This  Gull  is  merely  included  in  Favier's  list,  but  it  is  one  of 

the  most  abundant  species  of  Laridse  in  the  Straits  in  winter. 
The  greater  part  pass  northward  by  the  end  of  March  ;  but  some 
few  pairs  remain  to  nest  on  the  rocks  of  the  African  shore,  laying 
about  the  end  of  April. 

Head  and  neck  white  ;  mantle  slaty  black ;  legs  and  feet  yellow  ; 
primaries  nearly  uniform  dark  brown. 

Young.  Like  last.     Length  21-23  inches. 

346.  Larus   marinus,    Linnseus.     The  Great  Black -backed 

Gull. 

This  large  Gull  is,  according  to  Favier,  found  about  the  Straits 
in  small  numbers  from  January  to  March,  and  he  further  states 
that  he  never  saw  any  but  immature  birds ;  this  agrees  with  my 
own  observations,  except  that  I  have  occasionally  seen  adults. 

Resembles  the  last,  but  larger.  Legs  and  feet  flesh-colour  ;  distinct 
wedge  on  third  and  fourth  primaries. 

Young.  Like  last.     Length  26-33  inches. 

347.  Larus  glaucus,  0.  Fabricius.    The  Glaucous  Gull. 

This  Arctic  Gull  is  not  mentioned  by  Favier,  but  was  once 
obtained  by  him  in  immature  plumage  near  Tangier. 

Adult.  Entirely  wJiite,  except  grey  mantle  ;  head  and  neck  streaked 
wTith  ash-grey  in  winter. 

Young.  Dull  white,  mottled  with  pale  brown  ;  primaries  greyish  white  ; 
for  a  short  time  before  assuming  adult  plumage  the  mantle  is  entirely 
white.  Length  26-33  inches. 

348.  Rissa  tridactyla  (Linnseu*).    The  Kittiwake. 

"  This  species  is  nearly  as  common  during  winter  in  the  Straits 
as  the  Herring-Gull,  appearing  during  the  month  of  November 
and  leaving  in  March." — Favier. 

The  Kittiwake  is  to  be  seen  in  the  Bay  of  Gibraltar  during 
winter,  sometimes  in  great  abundance,  at  other  times  hardly  any 
are  to  be  found  ;  their  presence  or  absence  is  due  to  the  state  of 


302 

the  weather.  They  are  often  seen  about  the  Laguna  de  la  Janda, 
where,  in  February,  1892,  Verner  reports  seeing  vast  numbers  and 
caught  two  alive  so  gorged  with  black  worms  as  to  be  unable  to 
fly.  He  examined  one  and  found  a  round  mass  of  these  worms, 
of  the  size  of  a  golf-ball,  in  its  throat. 

Hind  toe  absent  or  rudimentary. 

In  summer.  Head,  neck,  and  tail  pure  white  ;  bill  greenish  yellow. 
In  winter.  Head  and  neck  slate-grey. 

Young.  Bill  black;  brown  feathers  on  back;  tail  with  black  band  at  end. 
Length  15  inches. 


Subfamily  STERCORARIIN.E.     ROBBER-GULLS. 

Base  of  upper  mandible  covered  with  a  cere  or  skin,  and 
hooked  at  the  point. 

349.  Stercorarius  catarractes  (Linnseus).    The  Great  Skua. 
Favier  records  a  single   specimen  obtained  near  Tangier,  in 

December,  1852.     It  occurs  regularly,  but  not  commonly,  during 
winter  in  the  Straits. 

General  colour  above  dark  brown;  below  lighter  brown;  axillaries  dark 
brown;  primaries  blackish  brown,  bases  white,  forming  a  well-marked 
band.  Tail  dark  brown,  white  at  the  base,  but  this  is  hidden  by  the 
coverts;  two  centre  feathers  very  slightly  longer  than  the  others.  Length 
24-25  inches,  leg  2f-3±. 

350.  Stercorarius   pomatorhinus  (Temminck).      The   Twist- 

tailed  or  Pomatorhine  Skua. 

Stated  by  Favier  to  be  very  rare  near  Tangier ;  and  he  only 
mentions  one  specimen,  obtained  as  far  back  as  November,  1845. 
Is  recorded  from  Malaga. 

Two  forms  or  races  are  found,  both  dark  brown  above.  The  dark  form 
as  dark  below  as  above;  the  light  form  white  below,  barred  on  flanks  and 
tail-coverts  with  dark  brown,  white  on  the  throat,  tinged  with  golden 
yellow,  and  sometimes  extending  round  nape. 

Two  centre  tail-feathers  4  inches  longer  than  the  others,  and  twisted 
upwards. 


STERCORARIUS  PARASITICUS.  303 

Young.  Above  dark  brown;  below  paler,  with  buff  margins  to  the 
feathers  ;  tail  brown,  two  centre  feathers  only  very  slightly  longer  than 
the  others.  Length  21  inches,  leg  2. 

351.  Stercorarius  crepidatus  (J.  F.  Gmelin).     Richardson's 

Skua. 

Spanish.  Cagalo  (Malaga)  (Arevalo). 

Favier  only  mentions  one  specimen  of  this  Skua,  killed  near 
Tangier  in  1844.  It  is,  however,  not  uncommon  in  winter.  "NY as 
often  seen  off  Malaga  in  March,  1882. 

Two  forms  or  races  are  found. 

Dark  race.  Adult  uniform  sooty  brown,  to  which  the  name  Richardson's 
is  strictly  applied;  but  the  name  Arctic,  also  employed,  is  a  misnomer,  as 
this  bird  breeds  and  also  migrates  further  south  than  any  of  the  other 
northern  Skuas  *. 

Light  race.  Adult  nearly  as  those  of  next  species,  but  two  centre  tail- 
feathers  only  3  inches  longer  than  the  others.  Legs  black. 

Young.  Sooty  brown;  paler  below;  centre  tail-feathers  little  longer  than 
the  others. 

All  primaries  with  white  shafts. 

Length,  including  centre  tail-feathers,  about  20  inches. 

352.  Stercorarius  parasiticus  (Linnseus).    The  Long- tailed  or 

Buffbn's  Skua. 

This  Skua  is  recorded  by  Favier  as  twice  obtained  near  Tangier 
— in  1846,  and  in  October,  1858,  the  first  being  an  immature 
specimen. 

Adult.  Crown  black;  above  brownish  grey;  below  white;  sides  of  neck 
white,  tinged  with  yellow;  legs  slate-grey.  Two  central  tail-feathers  about 
9  inches  longer  than  the  others. 

Young.  Sooty  brown;  flanks  and  tail-coverts  with  buff  margins;  centre 
tail-feathers  little  longer  than  the  others. 

First  two  primaries  with  white  shafts,  but  it  is  said  that  the  nostrils  of 
Buffon's  Skua  are  nearer  the  frontal  feathers  than  the  tip  of  the  bill — in 
Richardson's  Skua  the  contrary  being  the  case. 

Length,  including  central  tail-feathers,  22£  inches. 

*  The  writer  has  observed  a  light  male  and  a  dark  female  paired  together  at 
their  nest  on  one  of  the  Hebrides,  where  the  dark  and  light  forms  were  about 
equal  in  numbers. 


304  ALCID^E. 


Order   PYGOPODES.          Family  ALCID^ffiJ. 

353.  Alca  torda,  Linnseus.    The  Razor-bill. 

Moorish.  Bou-drihima  (Favier).     Spanish.  Gallareta  de  mar. 

Favier  only  says  of  this  species  that  it  "is  found  near  Tangier 
from  November  to  February." 

The  Razor-bill,  in  some  winters,  appears  in  the  Straits  in  very 
large  numbers,  as  in  the  winter  of  1871-72,  when,  during 
February,  they  were  to  be  seen  in  all  directions  about  Gibraltar 
Bay,  some  coming  into  the  New  Mole  so  close  to  the  land  that 
we  threw  stones  at  them,  and  they  lingered  on  very  late,  as  I  saw 
ten  on  the  19th,  one  on  the  21st,  and  two  on  the  28th  of  March, 
and  one  on  the  7th  of  April.  In  this  case  their  appearance  was, 
no  doubt,  attributable  in  the  first  instance  to  heavy  gales  and 
storms  outside  the  Straits. 

In  1877  I  saw  one  at  the  New  Mole  during  the  last  week  in 
May.  In  March,  1882,  there  were  vast  numbers  off  Malaga. 

Bill  black,  straight,  large,  much  decurved  towards  point,  with  curved 
white  line  across  centre  on  each  side;  upper  mandible  hooked,  with  three 
grooves  across. 

In  summer.  Above  black,  with  green  gloss  ;  secondaries  tipped  with 
white,  forming  a  band;  narrow  white  streak  from  base  of  upper  mandible 
to  eye;  below  white,  but  chin  and  throat  blackish. 

In  ivinter.  Green  gloss  above  nearly  absent ;  chin,  throat,  and  sides  of 
head  white;  in  young  birds  bill  less  developed.  Length  17  inches. 

354.  Uria  troile  (Linnseus).    The  Common  Guillemot. 

This  Guillemot  is  occasionally  seen  in  small  numbers  about 
the  Straits  in  winter,  especially  after  severe  weather  from  the 
westward. 

Bill  blackish,  straight  and  pointed. 

In  summer.  Head,  neck,  and  upper  surface  dark  brown;  below  white. 
In  winter.  Throat  and  sides  of  head  white.     Length  18  inches. 
The  Ringed  Guillemot  is  a  race  or  variety,  with  a  white  ring  round  the 
eye  and  a  white  line  running  backwards  therefrom. 


COLYMB1D.E.  305 

355.  Fratercula  arctica  (Linnaeus).    The  Puffin. 

Spanish.  Frailecillo. 

"  This  species  is  found  near  Tangier  from  November  to  March, 
sometimes  even  lingering  as  late  in  the  spring  as  April  and  May. 
They  are  more  abundant  than  the  Gannet,  and  are  frequently 
picked  up  dead  on  the  sea-shore  after  stormy  weather." — Favier. 

I  have  seen  Puffins  in  Gibraltar  Bay  as  late  as  the  5th  of 
March,  but  never  in  such  numbers  as  the  Razor-bill. 

Bill  shorter  than  head,  higher  than  long;  both  mandibles  arched  (and  in 
summer  grooved  across  with  orange). 

Above  black  ;  forehead  and  crown  greyish  brown ;  black  ring  round 
white  throat;  below  white;  axillaries  brown;  legs  orange. 

In  winter  the  bill  is  smaller  and  without  brighter  colour,  the  base  of  the 
bill  being  shed  in  autumn.  Length  12  inches. 


COLYMBIDJE.     Females  smaller  than  males. 

356.  Colymbus   glacialis,  Linnaeus.     The    Great   Northern 

Diver. 

Occasionally  seen  in  winter  in  the  Straits. 

Adult:  spring.  Head  and  neck  black,  with  purple  gloss;  a  band  of  white 
stripes  on  throat,  below  which  is  another  band  of  white  stripes;  bill  black; 
iris  red. 

Winter.  Above  uniform  blackish  brown,  more  or  less  spotted  with  white  ; 
below  white.  Length  30-33  inches. 

357.  Colymbus  arcticus,  Linnseus.  The  Black-throated  Diver. 
Occasionally  seen  in  winter  in  the  Straits. 

Adult :  spring.  Chin  and  throat  Hack,  with  a  purple  gloss,  divided  by  a 
patch  of  white  with  black  lines;  iris  red. 

Winter.  Like  last,  but  smaller.     Length  25  inches. 

358.  Colymbus  septentrionalis,  Linnaeus.    The  Red-throated 

Diver. 

This  Diver  is  common  in  the  Straits  in  winter,  and  I  obtained 
one  specimen  with  a  red  throat. 

x 


306 

Adult :  spring.    Throat  grey,  with  lower  part  chestnut-red. 
Winter.  Above  ashy  grey,  speckled  and  spotted  with  white,  most  so  in 
the  young;  underparts  white.     Length  21-23  inches. 


PODICIPEDID^].     Feet  with  lobes  ;  no  true  tail. 

359.    Podicipes    cristatus    (Linnseus).      The    Great    Crested 
Grebe. 

Spanish.  Zambullidor  (applied  to  all  Grebes  and  Divers). 

"  This  large  Grebe  is,  near  Tangier,  less  common  than  Podiceps 
minor.  Some  remain  in  the  country  to  breed ;  the  others  pass 
north  during  March.  They  are  very  abundant  at  the  lakes  of 
Has  el  Doura." — Favier. 

"We  can  quite  corroborate  the  latter  statement;  for  when  we 
were  at  those  lakes  at  the  end  of  April,  the  number  of  these 
Grebes,  as  well  as  of  the  next  species,  was  perfectly  marvellous. 
They  were  in  pairs,  but  had  not  commenced  laying.  These 
swampy  lakes,  much  covered  at  the  sides  with  aquatic  plants 
and  sedges,  must  be  a  paradise  for  all  Grebes  and  water-birds ; 
but  it  is  vexation  of  spirit  and  almost  useless  for  the  ornithologist 
to  go  there.  The  Arabs,  at  the  egging-season,  move  their  tents 
close  to  the  lake  and  plunder  every  nest  they  can  find,  and 
further  pester  Europeans  to  an  unbearable  degree,  being  almost 
as  annoying  and  intrusive  as  the  mosquitoes,  which  were  there 
more  troublesome  than  in  any  country  I  have  been  in.  Towards 
Casas  Viejas  and  Gibraltar  I  never  obtained  the  present  species ; 
but  Lord  Lilford  found  them  breeding  plentifully  in  May  near 
the  edge  of  the  Goto  de  Donana. 

Male  :  summer.  Crown  and  tufts  black  ;  tippet  round  face  chestnut, 
with  black  edges. 

Female.  Rather  smaller  ;  crest  and  tippet  not  so  much  developed. 

In  winter  these  ornaments  are  scarcely  present  in  the  adult  male,  less  so 
in  the  female,  and  absent  in  the  young  ;  all  have  white  eye-stripe.  Length 
20-24  inches. 


PODICIPES  NIGRICOLLIS.  307 

360.  Podicipes  grriseig;ena(Boddaevt).  The  Red-necked  Grebe. 
"  This  bird  is  less  common  near  Tangier  than  the  last,  being 

seldom  observed  on  passage.  Some  remain  in  the  country  to 
breed,  the  others  migrating  northwards  in  March,  returning 
again  during  September.  They  are  more  abundant  at  the  lakes 
of  Ras  el  Doura,  and  are  there  called  Mazan  by  the  Arabs." — 
Favier. 

Some  specimens  of  the  Red-necked  Grebe  obtained  in  Morocco 
by  Favier  were  so  young  that  they  must  have  been  bred  in  the 
country ;  and  we  saw  many  at  the  above  lakes  in  April,  but  I 
have  no  record  of  them  on  the  Spanish  side. 

In  summer.  Crown,  nape,  back  of  neck  black  ;  chin   and  throat  grey  ; 
neck  in  front  chestnut-red  ;  no  tippet ;  bill  black,  base  of  bill  yellow. 
In  winter.  Neck  in  front  grey  ;  no  eye-stripe.     Length  16^  inches. 

361.  Podicipes    auritus  (Linnaeus).     The   Lesser  Crested  or 

Slavonian  Grebe. 

Although  Favier  has  not  mentioned  this  Grebe  as  occurring 
near  Tangier,  I  have  seen  one  specimen  obtained  in  the  Straits 
in  October,  1867 ;  it  probably  is  often  to  be  met  with  in  winter. 

Bill  straight. 

In  summer.  With  tuft  of  lengthened  chestnut  feathers  on  each  side  of 
head  from  bill  to  nape  ;  crown,  forehead,  chin,  and  ruff  or  tippet  black  ; 
neck  and  flanks  chestnut. 

In  winter.  As  in  last.     Length  13  inches. 

362.  Podicipes  nigricollis,  C.  L.  Brehm.    The  Black-necked 

or  Eared  Grebe. 

This  species  is  the  most  common  of  the  Grebes,  breeding  in 
lagoons  and  swamps  on  both  sides  of  the  Straits.  In  the  winter 
they  take  to  the  salt  water,  and  are  generally  plentiful  in 
Gibraltar  Bay. 

Bill  slightly  curved  upwards  ;  inner  four  primaries  white  throughout. 

In  summer.  Head  and  neck  black,  with  a  stripe  from  eye  to  nape  of 
lengthened  golden-reddish  feathers. 

In  winter.  Resembles  last  in  plumage,  except  the  white  on  primaries. 
Length  12  inches. 

x2 


308  PEOCELLAR11D.E. 

363.  Podicipes   fluviatilis  (Tunstall).     The  Little  Grebe    or 

Dabchick. 

Moorish.  El  ghotis  (Favier). 

"  This  small  Grebe  is  resident  near  Tangier,  although  to  a  great 
extent  migratory,  passing  north  during  April,  and  reappearing 
from  October  to  December.  They  are  resident  and  especially 
numerous  at  the  lakes  of  Kas  el  Doura,  where  the  Arabs,  during 
the  breeding-season,  in  a  great  measure  subsist  on  the  eggs  of 
various  aquatic  birds,  destroying  a  prodigious  quantity." — Favier. 

The  Dabchick  is  resident  in  Andalucia,  breeding  abundantly 
in  some  localities  ;  but  is  most  common  in  winter  ;  and  how  they 
reach  the  isolated  patches  of  water,  which  are  dry  in  summer,  is 
marvellous,  as  I  never  saw  one  on  the  wing. 

The  Dabchick,  in  winter,  is  almost  always  to  be  seen  on  the 
.inundation  at  the  North  Front  of  Gibraltar,  taking  no  notice  of 
the  numerous  passers  by. 

In  summer.  Chin  black;  cheeks,  sides,  and  front  of  neck  reddish  chestnut. 
In  winter.  Chin  white,  the  reddish  chestnut  of  neck  replaced  by  brown  ; 
iris  brown.     Length  8-9  inches. 

Order  TUBINARES. 

Family  PROCELLARIIDJE.     Nostrils  in  a  tube. 
Oceanic  birds,  seldom  landing  except  to  breed. 

364.  Puffinus  kuhli,  Boie.    The  Cinereous  Shearwater. 
This  species  of  Shearwater  is  abundant  in  the  Straits,  and  is 

occasionally  found  dead  on  the  shore.  They  nest  about  the 
beginning  of  May,  under  rocks  and  stones  on  islands,  and  a 
Shearwater  is  reported  to  breed  at  the  back  of  the  Eock. 

Head,  nape,  and  back  dull  brown  ;  cheeks  and  chin  grey  ;  below  entirely 
white. 

Bill  yellowish,  tips  horn-colour  ;  legs  flesh-colour  (10th  March).  Length 
18  inches. 

Puffinus  gravis,  likely  to  occur  in  the  Straits,  is  similar  in  size,  but  lias 
head  and  nape  dark  brown,  forming  a  cap  ;  cheeks  and  chin  white  ;  centre 
of  belly  brownish.  =P.  major }  auctorum. 


PROCELLARIA  PELAGICA.  309 

365.  Pufflnus  anglorum  (Temminck).    The  Manx  Shearwater. 

Spanish.  Animas  (souls),  Diablos  (devils)  *. 

Favier  states  that  this  Shearwater  is  "  found  from  August  to 
November,  and  usually  picked  up  dead  on  the  sea-shore." 

It  is  common  in  the  Straits  in  autumn,  occasionally  coming 
close  in  to  the  land  in  the  Bay  of  Gibraltar. 

They  seem  to  feed  where  porpoises  swim  at  the  surface  of  the 
water,  and  were  thus  seen  in  flocks  of  forty  and  fifty  off  Malaga 
in  March,  1882,  often  sitting  together  on  the  water  in  company 
with  P.  Jculdi.  We  shot  a  great  many  of  both  species,  which 
were  eaten  by  the  yacht's  crew. 

Mediterranean  specimens  have  the  back  lighter  than  northern 
birds,  and  are  marked  with  grey  on  the  flanks.  This  climatic 
race  has  been  named  P.  yelkouan,  Acerbi. 

Head,  back,  wings,  and  tail  blackish  brown  ;  breast  and  flanks  white  ; 
rest  of  underparts  dark  smoky  grey  ;  bill  and  legs  bluish  grey.  Length 
13-14  inches,  wing  9^. 

366.  Puffinus  obscurus   (J.  F.  Gmelin).    The  Dusky  Shear- 

water. 

Recorded  from  Malaga. 

Smaller  than  Manx  Shearwater.  Above  blackish  brown  ;  underparts 
pure  white  ;  axillaries  white.  Length  11  inches. 

367.  Procellaria  pelagica,  Linnaeus.    The  Storm-Petrel. 
This  little  Petrel  is  frequently  seen  skimming  about  in  the 

Straits,  and,  no  doubt,  nests  on  some  of  tbe  small  islands  or 
patches  of  rock  on  the  coast,  as  they  are  reported  to  do  near 
Cabo  de  Palos,  Mar  Menor. 

Above  slaty  black  ;  upper  tail-coverts  and  sides  of  vent  white  ;  below 
darker  ;  tail  slightly  rounded  ;  legs  and  feet  black.  Length  5£  inches. 
Smallest  web-footed  bird. 

*  This  Shearwater  is  very  numerous  in  the  Bosphorus  and  Dardanelles,  passing 
continuously  up  and  down  in  small  flights,  never  being  seen  except  on  the  wing  ; 
they  are  known  as  the  "  souls  of  the  damned,"  who  are  allowed  no  rest. 


310 


PROCELLA.RIID.E. 


368.  Procellaria  leucorrhoa,  Vieillot.    Fork-tailed  Petrel. 
This  species  is  stated  by  Favier  to  be  of  rare  occurrence  in  the 

Straits.     Those  which  he  obtained  were  all  found  dead   on  the 
sea  shore  after  storms.     He  mentions  picking  up  six  in  1846,  and 
one  in  each  of  the  years  1852,  1854,  and  1858. 
Is  recorded  from  Malaga. 

Slate-black  above,  darker  below  ;  upper  tail-coverts  white  ;  tail  long, 
deeply  forked.  Length  7£  inches. 

369.  Oceanites    oceanicus    (Kuhl).      The    Long-legged    or 

Wilson's  Petrel. 

This  wide-ranging  Petrel  was  obtained  near  Malaga  by  Fran- 
cisco de  los  Rios. 

Sooty  black,  darkest  below  ;  upper  tail-coverts  and  sides  of  vent  white  ; 
tail  almost  square  ;  hind  toe  absent. 

Legs  \\  inches  long  and  slender,  covered  with  large  scales  ;  toes  black, 
with  the  centre  of  the  webs  yellow.  Length  1\  inches. 


LAJA.    DEL    CISCA.B. 


APPENDIX. 


THE  following  list  of  Lepidoptera  may  be  of  use  to  collectors, 
and  is  chiefly  compiled  from  "  Notes  on  Lepidoptera  from  the 
Region  of  the  Straits  of  Gibraltar,"  by  James  J.  Walker,  R.N., 
F.L.S.,  published  in  the  *  Transactions  of  the  Entomological 
Society  of  London,'  1890,'pp.  361-391. 

Most  of  the  Butterflies  have  been  obtained  by  my  son,  L.  P. 
Irby,  60th  Rifles,  and  myself;  possibly  a  few  more  species  may 
be  found. 

The  list  of  Moths  is  incomplete,  and  is  almost  entirely  from 
Mr.  Walker's  paper,  to  whom  I  am  much  indebted  for  permission 
to  make  use  of  his  notes  and  for  his  kind  assistance  in  looking 
over  this  list. 

Those  species  marked  with  an  asterisk  occur  on  the  Rock  of 
Gibraltar.  All  the  Moorish  specimens  were  found  by  Mr.  Walker. 

RHOPALOCEBA  (BUTTEEFLIES). 

*Papilio  podalinus.  L.     Gibraltar  (L.  P.  Irby). 

P.  machaon,  L.     February  to  October. 
*  Thais  rumina,  L.     March,  April. 
*Pieris  brassicce,  L.     All  the  year. 
*P.  rapce,  L.     All  the  year. 

P.  napi,  L.     Cork-wood  (L.  P.  Irby). 
*P.  daplidice,  L.     March  to  November. 
*Euchloe  belemia,  Esp.     January,  February  (ab.  glauce,  Hb.). 

E.  belia,  Esp.     February. 

E.  tagis,  Esp.     March. 

E.  cardamines,  L.     Cork-wood,  18th  April,  1894. 


312  APPENDIX. 

*Euchloe  eupJienoides,  Stand.     March,  April. 

E.  eupheno,  L.     Morocco  only.     March. 

LeucopJiasia  sinctpis,  L.     March  to  July. 
*Colias  edusa,  F.,  and  ab.  Tielice,  Hiibn.     All  the  year. 

Gonopteryx  rhamni,  L.     (Hibernates.)     June. 
*6r.  cleopatra,  L.     (Hibernates.)     May,  June,  July. 
*Theda  spini,  Schiff.,  and  ab.  $  lynceus,  Hiibn. 

T.  ilicis,  Esp.,  v.  cesculi,  O.     May,  Jnne. 

T.  quercus,  L.     July. 

T.  roboris,  Esp.     June. 

T.  rubi,  L.     February  and  March,  in  numbers. 
*Thestor  baling,  Hiibn.     February  to  April. 

T.  mauritanicus,  Lucas.     Morocco  only.     March,  April. 

Polyommatus  pTdceas,  L.     All  the  year  round,  and  v.  eleus,  F.,  in  summer. 
*Lijccena  bcetica,  L.     All  the  year. 

L.  telicanus,  Hiibn.     March  to  October. 

L.  iheopTirastus,  F.     Only  near  Tetuan.     July. 

L.  baton,  Berg,  v.  panoptes,  Hiibn.     May. 

L.  lysimon,  Hiibn.     August,  November. 

L.  astrarche,  Berg.     All  the  year. 
*L.  icarus,  Eott.     February  to  November. 

L.  bellargus,  Eott.     April  and  July. 

L.  minimus,  Fuess.,  v.  lorquinii,  H.-S.     May  16th. 

L.  melanops,  Bdv.     April,  May. 
*L.  argiolu-s,  L.     January  to  July. 

*Charaxasjasius,Ij.     Bare  at  Gibraltar;  one  in  1872.     Common  at  Tangier 
in  September. 

*  Vanessa  polychloros,  L.     (Hibernates.)     June. 

V.  urticce,  L.     (L.  P.  Irby.)     Also  included  by  Bosenhauer. 

*  V.  atalanta,  L.     All  the  year. 

*  V.  cardui,  L.     All  the  year. 

Melitcea  aurinia,  Bott.,  v.  desfontainii,  Godt.     May  27th. 

M.  phoebe,  Kn.,  v.  cttheria,  Hiibn.     May. 

Argynnis  latona,  Esp.     June. 

*A.  pandora,  Schiff.     May  to  July  ;  September,  Gibraltar. 
*Danaisplexippus,  L.     One,  Gibraltar,  Oct.  24,  1886  (J.  J.  Walke. 

Melanargia  ines,  Hfsgg.  (tJietis,  Hiibn.).     April,  May. 
*Satyrusfidia,  L.     June  to  August. 
*S.  statilimis,  Hufn.,  v.  allionia,  F.     July,  August. 

Pararge  mcera,  L.,  v.  adraste,  Hiibn.     April. 
*P.  megcera,  L.     January  to  October. 
*P.  cegeria,  L.     All  the  year. 

*EpinepJiele  janira,  L.,  v.  hispulla,  Hiibn.     May  to  August. 
*E.  ida,  Esp.     May  to  September. 


APPENDIX.  313 

*Epinephele  pasiphae,  Esp.     April  to  June. 

Ccenonympha  arcanioides,  Pierret.     Morocco  only.     March. 

C.  dorus,  Esp.     July. 
*C.  pamphilus,  L.     May,  June. 
*Spilotliyrus  alcece,  Esp.     May,  June. 

S.  althece,  Hiibn.     All  the  year. 

Syriclithus  proto,  Esp.     May,  July. 

S.  fritillum,  Hiibn.     July. 

S.  sao,  Hiibn.,  v.  thempne,  Ebr.     April,  July. 

Hespena  thaumas,  Hufn.     May,  June. 

II.  action,  Esp.     May,  June. 
*H.  nostradamus,  F.     June  to  September. 

H.  zelleri,  Lederer.     Morocco  only.     July  to  November. 

HETEROCEKA  (Moras),  to  end  of  GEOMETER. 
SPHINGES. 

*Acherontia  atropos,  L.     Many  larvae  on  Solanum  sodomceum  (L.  P.  Irby,  1894). 

*Sphinx  convolvuli,  L. 

*Dci1ephila  eupJiorbice,  L.     Many  larvae  (L.  P.  Irby,  1894). 

*D.  livornica,  L.     Summer. 

*  Chcerocampa  celerio,  L.     September,  October. 

Smerinthus  populi,  L.     Morocco. 

Pterogon  proserpina,  Pall.     Larvae  common  on  (Enothera,  June  (J.  J.  AValker). 
* Macroglossa  stettatarum,  L.     All  the  year.  k 

*Sdapteron  tabaniforme,  Kott.     July. 

Sesia  ramburi,  Stand.     June. 
*Paranthrene  tineiformis,  Esp.     June. 

Zygcena  sarpedon,  Hiibn.     May,  June. 

Z.  st&chadis,  Bork.     June. 

Z.  lavandulce,  Esp.     May. 

Z.  bcetica,  Ebr.     May,  August,  September. 

BOMBTCES. 

*Nola  cicatricalis,  Tr.     April,  October. 

*Emydia  cribrum,  L.,  v.  Candida,  Cyrill.     May,  August,  September. 

*Deiopeia  pulchella,  L.     All  the  year. 

Euchelia  jacobcece,  L.     June. 
*Euprepria  pudica,  L.     August,  September. 

Arctia  villica,  L.     March,  April. 

A.  casta,  Esp.     April. 

Phmymatobia  fuliginosa,  L.     Esmir,  Morocco. 
*Porthesia  auriflua,  L.     June. 


314  APPENDIX. 

Orgyia  Josephines,  Oberth,     Esmir,  Morocco. 
*0.  trigotephras,  Bdv.     June. 

*0cneria  dispar,  L.     June.     Larvae  in  astonishing  abundance  near   Algeciraz, 
June  1888  (J.  J.  Walker). 

0.  ailantica,  Rbr. 

Cossus  ligniperda,  L. 
*Zeuzera  pyrma,  L.     June. 
*Bombyx  neustria,  L.     September. 
*B.  trifolii,  L.     June. 
*Megasoma  repanda,  Hiibn.     August. 

Saturnia  pavonia,  L.     April. 
*S.pyri,  L.     March. 

Cerura  vinula,  L.     Tangier,  May. 
*C.  bifida,  L.     August. 

Pygcera  bucepTiala,  L.     March. 


*Ra<phia  hybris,  Hiibn.     July. 

Acronycta  psi,  L.     Morocco,  April. 

A.  rumiris,  L.     Morocco,  February. 
*Bryophila  muralis,  Forst.,  v.  par,  Hiibn.     July. 
*B.  ereptricula,  Tr.     August,  September. 

Tapinostola  musculosa,  L.     May. 
*Leucania  loreyi,  Dup.     October. 
*L,  l-album,  L.     October,  November. 
*Caradrina  exigua,  Hiibn.     October. 
*C.  ambigua,  W.  V.     April,  October. 

Pachnobia  rubricosa,  "W  .  V.     May. 
*0rrhodla  erythrocepTiala,  W.  V.     October. 
*Triphcena  pronuba,  L.     Summer. 
*Agrotis  dahlii,  Hiibn.     November. 
*A.  saucia,  Hiibn.     All  the  year. 
*A.  hucoc/aster,  Frr.     One,  Tetuan. 
*A.  puta,  Hiibn.     October. 
*A.  segetum,  SchifE.     October,  November. 

A.  spinifera,  Hiibn.     February,  October. 
*A.  ypsilon,  Eott.,  suffusa,  Hiibn.     October. 
*A.  crassa,  Hiibn.     September. 

*Briihys  pancratii,    Cyr.      February,    November.       Larvae    on    Pancratium 
mauritimum. 

Dryobata  saportce,  Dup.     October,  November. 
*Mamestra  serena,  W.  V.     March. 
*Jf.  dysodea,  W.  V.,  chrysozona,  Bork.     April. 
*M.  oleracea,  L.     October. 


APPENDIX,  315 

*Triyouopliomflamnua,  Esp.  (empyrea,  Hiibn.). 
*Eriopus  latre'dlei,  Dup.     October. 
*Calocampa  vetusta,  Hiibn.     October. 

Calophasia  platyptera,  Esp.     Tangier,  September. 

Cleophana  antirrJiini,  Htibn.     May. 

C.  yvanii,  Dup.     May. 
*Cucullia  verbasci,  L.     April. 
*C.  chamomillce,  W.  V. 
*Eurhip!a  adulatrix,  Bdv.     September. 

Heliothis  armlgera,  Hiibn.     October. 

//.  incarnata,  Fr.     May,  July. 
*H.  peltigera,  W.  V.     May  to  July. 
*PLusia  chalcites,  Esp.     October. 
*P.  gamma,  L.     All  the  year. 

Acontia  luctuosa,  W.  V.     May  to  September. 

A.  lucida,  Hiibn.,  v.  albicollis,  JF.     Sierra  Eetin,  March  28,  1894. 
*Catocala  elocata,  Esp.     September. 

O.  dilecta,  Hiibn.     July. 

C.  promissa,  Esp.     June. 

(7.  conversa,  Esp.     June. 

C.  nymphagoga,  Esp.     June. 

Cerocala  scapulosa,  Bdv.     April,  May. 

Ophiusa  bifasciata,  Petagna.     Tangier,  September  1888. 

0.  algira,  L. 

Pseudophia  lunaris,  W.  V.     March. 
*P.  tirrhcea,  Cr.     Larvae  on  Schinus  molle, 

Spiniherops  spectrum,  F.     June. 

Agrophila  trabealis,  Scop.     August. 

Prothymia  conicephala,  Staud.     July. 
*  Thalpochares  ostrina,  Hiibn.     April  to  October. 
*T.  parva,  Hiibn.     August. 
*Metophia  monogramma,  Hiibn.     April,  May. 
*Herminia  crinalis,  Tr.     October. 
*H.  lividalis,  Hiibn.     June,  October. 
*H.  obsitalis,  Hiibn.     October. 

GEOMETER. 

*Nemoria  herbaria,  Hiibn.     October. 

Acidalia  vittaria,  Hiibn.     April,  June. 

A.  ochrata,  Scop.     June. 

A.  perochraria,  Eossl.     June. 

A.  nexata,  Hiibn.     June,  October. 
*A.  virgularia,  Hiibn.     October. 


316  APPENDIX. 

Acidalia  elongaria,  Rbr.     May. 

A.  circuitaria,  Hiibn.     Morocco  only.     June. 

A.  ostrinaria,  Hiibn.     June. 

A.  ?  transmutaria,  Rbr.     June. 

A.  promutata,  Guen.  (marginepunctata,  Goze).     July. 

A.  luridata,  Zell.     September. 

A.  emutaria,  Hiibn.     Esmir,  Morocco.     November. 

A.  imitaria,  Hiibn.     July. 

A.  ornata,  Scop.     May  to  August. 

Zonosoma  pupillaria,  Hiibn.     July. 

Z.  porata,  F.     July. 

Pellonia  calabraria,  Zell.     April,  May. 

Abraxas  pantaria,  L.     April,  July. 

Terpnomicta  dilectaria,  Hiibn.     Tangier,  September. 

Boarmia  gemmaria,  Brahm  (rhotnboidaria,  W.  V.).     June. 

Pachychnemia  hippocastanaria,  Hiibn.     May. 
*Onophos  respersaria,  Hiibn.     June. 
*6r.  mucidaria,  Hiibn.     October. 

Anthometra  plumularia,  Bdv.     June. 

Fidonia  plumistaria,  Vill.     Sierra  Eetin,  26tb  March,  1894. 

Sdidosema  ericetaria,  Vill.     Summer  and  autumn. 

Thamnonoma  gesticularia,  Hiibn.     May,  June. 
*Aspilates  ochrearia,  Eossi.     April. 

Lozogramma  obtusaria,  Walker.      A  New-Zealand  species,  once  taken  near 
Gibraltar  by  Mr.  J.  J.  Walker,  19tb  March,  1888. 

Ligia  opacaria,  Hiibn.     October. 
*Sterrha  sacraria,  L.     All  the  year. 
*S.  consecraria,  Eamb.     Once  in  October. 
*Lithostege  griseata,  W.  V.     May. 

Anaitis  plagiata,  L.     May,  June. 

Chesias  rufata,  F.,  v.  obliquaria,  W.  V.     March. 

Ortholitha  peribolata,  Hiibn.     October. 
* Cidaria  fluctuata,  L.     Autumn. 
*C.Jluviata,  Hiibn.    All  the  year. 

C.  basochesiata,  Dup.     October. 

(7.  bilineata,  L.     July. 
*Eupithecia  centaureata,  W.  V.  (oblongata,  Thub.).     Spring  and  autumn. 

E.  satyrata,  Hiibn.    May,  June. 
*E.  pumilata,  Hiibn.     Spring  and  autumn. 


INDEX   TO    BIRDS. 


Accentor,  Alpine,  49. 
Accentor  collaris,  49. 

—  modularis,  49. 
Accipiter  nisus,  186. 
Acredula  caudata,  71. 

-  irbii,  71. 

—  rosea,  71. 
Acrocephalus  aquaticus,  58. 

—  palustris,  58. 

—  phragmitis,  58. 

—  streperus,  57. 

—  turdoides,  57. 
Aedon  galactodes,  56. 
-/Egialitis  cantiana,  270. 

—  curonica,  269. 

hiaticula,  268. 

.^Egithalus  pendulinus,  74. 
Alauda  arborea,  118. 
arvensis,  117. 

-  bsetica,  120. 

-  bracbydactyla,  120. 

—  calandra,  121. 

cristata,  118. 

macrorhyncha,  119. 

-  theckLe,  119. 
Alca  torda,  304. 
Alcedo  ispida,  130. 
Anas  boscas,  220. 
strepera,  221. 


Anser  albifrons,  217. 
brachyrhynchus,  218. 

—  cinereus,  216. 

—  erythropus,  217. 
segetum,  218. 

Anthus  campestris,  116. 
cervinus,  115. 

—  obscurus,  1 15. 
pratensis,  115. 

—  richardi,  116. 

—  spipolecta,  116. 

—  trivialis,  114. 
Aquila  adalberti,  171. 

chrysaetus,  170. 

na3via,  173. 

rapax,  174. 

Ardea  alba,  202. 

-  bubulcus,  203. 
cinerea,  200. 

-  garzetta,  202. 

-  purpurea,  200. 

-  ralloides,  204. 
Ardetta  minuta,  206. 
Argya  fulva,  70. 
Asio  accipitrinus,  139. 
capensis,  140. 

-  otus,  139. 

Astur  palumbarius,  185. 
Avocet,  274. 


318 


INDEX  TO  BIRDS. 


Babbler,  North-African,  70. 
Bee-eater,  131. 
Bernicla  leucopsis,  218. 
Bittern,  Common,  206. 

— ,  Little,  206. 
Blackbird,  35. 
Blackcap,  51. 
Bluethroat,  Red-spotted,  46. 

— ,  White-spotted,  46. 
Botaurus  stellaris,  206. 
Brambling,  104. 
Bubo  ignavus,  143. 
Bulbul,  White-vented,  67. 
Bullfinch,  Desert,  106. 
Bunting,  Cirl,  108. 

— ,  Corn,  107. 

— ,  House,  110. 

— ,  Little,  110. 

— ,  Ortolan,  109. 

— ,  "Reed,  109. 

— ,  Rock,  110. 

,  Snow,  111. 

Bustard,  Great,  253. 

— ,  Little,  257. 

— ,  North-African,  260. 

— ,  Ruffed,  260. 
Butcher-bird,  88. 
Buteo  desertorum,  169. 

—  vulgaris,  168. 
Buzzard,  Common,  168. 

— ,  Honey,  189. 
• ,  Rufous,  169. 

Caccabis  petrosa,  239. 

-  rufa,  237. 
Calidris  arenaria,  280. 
Caprimulgus  europa3us,  125. 

-  ruficollis,  126. 
Carduelis  elegans,  98. 

spinus,  100. 

Carine  glaux,  145. 

noctua,  144. 

Certhia  familiaris,  75. 
Certhilauda  bifasciata,  117. 
duponti,  117. 

—  lusitanica,  117. 


Cettia  cettii,  61. 
Chaffinch,  European,  102. 
,  North-African,  103. 


Charadrius  fulvus,  267. 
pluvialis,  266. 


Chelidon  urbica,  94. 
Chiffchaff,  63. 
Chough,  Red-billed,  80. 
,  Yellow-billed,  80. 


Chrysomitris  citrinella,  100. 
Ciconia  alba,  207. 
nigra,  210. 


Cinclus  albicollis,  76. 
aquaticus,  76. 


Circaetus  gallicus,  182. 
Circus  seruginosus,  163. 

cineraceus,  167. 

—  cyaneus,  166. 
swainsoni,  166. 


Cisticola  cursitans,  69. 
Clangula  glaucion,  229. 
Coccothraustes  vulgaris,  97 
Coccystes  glandarius,  136. 
Columba  livia,  234. 
— —  cenas,  233. 

—  palumbus,  232. 
Colymbus  arcticus,  305. 

glacialis,  305. 

septentrionalis,  305. 

Coot,  Common,  248. 

— ,  Red-lobed,  249. 
Coracias  garrula,  130. 
Cormorant,  198. 

— ,  Green,  198. 
Corn-Crake,  244. 
Corvus  corax,  84. 

comix,  86. 

corone,  86. 

frugilegus,  86. 

monedula,  83. 

tingitanus,  84. 

Cotile  riparia,  95. 

rupestris,  96. 

Coturnix  communis,  240. 
Courser,  Cream-coloured,  263. 
Crake,  Baillon's,  245. 


INDEX  TO  BIRDS. 


319 


Crake,  Corn,  244. 

,  Little,  245. 

,  Spotted,  244. 

Crane,  Common,  249. 

,  Demoiselle,  252. 

Creeper,  Rock,  75. 

,  Tree,  75. 

Crex  pratensis,  244. 
Crossbill,  Common,  106. 

— ,  Parrot,  107. 
Crow,  Carrion,  86. 
,  Grey,  86. 

— ,  Hooded,  86. 
Cuckoo,  Common,  135. 

— ,  Great  Spotted,  136. 
Cuculus  canorus,  135. 
Curlew,  Common,  290. 
,  Mediterranean,  290. 

— ,  Stone,  262. 
Cursorius  gallicus,  263. 
Cyanopica  cooki,  82. 
Cygnus  musicus,  215. 

-  olor,  216. 
Cypselus  apus,  122. 

—  melba,  124. 

—  murinns,  123. 

Dabchick,  308. 
Dafila  acuta,  222. 
Daulias  luscinia,  47. 
Dendrocopus  major,  126. 

mauritanus,  127. 

medius,  127. 

numidicus,  127. 

Dipper,  Greyrbacked,  76. 
Diver,  Black-throated,  305. 

,  Great  Northern,  305. 

,  Red-throated,  305. 

Dotterel,  268. 

Dove,  Common  Turtle,  234. 

— ,  Egyptian  Turtle,  235. 

— ,  Ring,  232. 
,  Rock,  234. 

— ,  Stock,  233. 
Duck,  Black,  229. 

— ,  Brahminy,  219. 


Duck,  Ferruginous,  227. 
,  Marbled,  223. 

— ,  Scaup,  229. 

— ,  Spiny-tailed,  230. 

— ,  Tufted,  228. 

,  White-faced,  230. 

,  Wild,  220. 

Dunlin,  281. 

Eagle,  BoneUi's,  174. 

— ,  Booted,  180. 

,  Golden,  170. 

,  Sea,  174. 

,  Short-toed,  182. 

— ,  Snake,  182. 

,  Spotted,  173. 

,  Tawny,  174. 

,  White-shouldered,  171. 

— ,  White-tailed,  174. 
Egret,  Little,  202. 
Elanus  caeruleus,  190. 
Emberiza  cia,  110. 

cirlus,  108. 

citrinella,  108. 

hortulana,  109. 

miliaria,  107. 

pusilla,  110. 

saharaB,  110. 

schceniclus,  109. 

Erismatura  leucocephala,  230. 
Erythacus  rubecula,  46. 
Erythrospiza  githaginea,  106. 
Eudromias  mormellus,  268. 
Eupodotis  arabs,  260. 

Falco  aesalon,  194. 
babylonicus,  192. 

—  barbarus,  192. 

cenchris,  196. 

eleonorae,  193. 

lanarius,  192. 

—  pcregrinus,  190. 

punicus,  191. 

subbuteo,  193. 

tinnunculus,  195. 


320 


INDEX  TO  BIRDS. 


Falco  vespertinus,  195. 
Falcon,  Barbary,  192. 

— ,  Eleonora,  193. 

— ,  Peregrine,  190. 

— ,  Mediterranean,  190. 

,  Rufous-naped,  192. 

Fieldfare,  35. 
Finch,  Citril,  100. 

— ,  Serin,  101. 

— ,  Snow,  104. 
Flamingo,  213. 
Flycatcher,  Collared,  92. 

— ,  Pied,  92. 

— ,  Spotted,  91. 

Francolin,  Double-spurred,  240. 
Francolinus  bicalearatus,  240. 
Fratercula  arctica,  305. 
Fringilla  coalebs,  102. 
incerta,  106. 

—  montifringilla,  104. 

-  spodiogenys,  103. 
Fulica  atra,  248. 

—  cristata,  249. 
Fuligula  cristata,  228. 

-  ferina,  226. 

-  mania,  229. 

—  nyroca,  227. 

-  rufila,  226. 

Gadwall,  221. 

Galerita  macrorhyncha,  119. 

-  thecklae,  119. 
Gallinago  brehmi,  279. 

cffilestis,  278. 

delamottii,  279. 

—  gallinula,  279. 

major,  277. 

russata,  279. 

Gallinula  chloropus,  247. 
Gannet,  199. 
Garganey,  225. 
Garrulus  cervicalis,  82. 

glandarius,  81. 

Gecinus  canus,  129. 

sharpii,  128. 

vaillanti,  128. 


Gecinus  viridis,  128. 
Glareola  pratincola,  261. 
Goatsucker,  125. 
Godwit,  Bar-taUed,  288. 

— ,  Black-tailed,  288. 
Golden -eye,  229. 
Goldfinch,  98. 
Goosander,  231. 
Goose,  Bean,  218. 

— ,  Bernicle,  218. 

— ,  Grey-lag,  216. 

,  Lesser  White-fronted,  217. 

,  Pink-footed,  218. 

— ,  Solan,  199. 

— ,  White-fronted,  217. 
Goshawk,  185. 
Grebe,  Black-necked,  307. 
,  Eared,  307. 

— ,  Great  Crested,  306. 

— ,  Lesser  Crested,  307. 

— ,  Little,  308. 
,  Red-necked,  307. 

— ,  Slavonian,  307. 
Greenfinch,  98. 
Greenshank,  286. 

,  Lesser,  286. 

Grosbeak,  Scarlet,  106. 
Grouse,  Black-bellied  Sand,  237. 

— ,  Pin-tailed  Sand,  236. 
Grus  communis,  249. 

—  virgo,  252. 
Guillemot,  Common,  304. 
Gull,  Adriatic  Black-headed,  297. 

— ,  Audouin's,  299. 
,  Brown-headed,  297. 

— ,  Common,  298. 
,  Glaucous,  301. 

— ,  Great  Black-backed,  301. 

— ,  Herring,  300. 
,  Lesser  Black-backed,  301. 

— ,  Little,  297. 

— ,  Slender-billed,  299. 

,  Southern,  300. 

-,  Yellow-legged  Herring,  300. 


Gypaetus  barbatus,  156. 
Gyps  fulvus,  147. 


INDEX  TO  BIRDS. 


121 


Haomatopus  ostralegus,  273. 
Haliaetus  albicilla,  174. 
Hammer,  Yellow,  108. 
Harrier,  Hen,  166. 
— ,  Marsh,  163. 

,  Montagu's,  167. 

,  Pale-chested,  166. 

Hawfinch,  97. 

Hawk,  Black-shouldered,  190. 

,  Many-banded,  168. 

Hedge-Sparrow,  49. 
Heron,  Buff-backed,  203. 
,  Cattle,  203. 


-,  Common,  200. 
-,  Great  White,  202. 
-,  Night,  205. 
-,  Purple,  200. 
-,  Squacco,  204. 


Himantopus  candidus,  274. 
Hirundo  rufula,  94. 

rustica,  93. 

Hobby,  193. 

.,  Western  Eed-footed,  195. 

Hoopoe,  134. 
Houbara  undulata,  260. 
Hydrochelidon  hybrida,  292. 

leucoptera,  292. 

nigra,  291. 

Hypolais  icterina,  65. 

olivetorum,  66. 

opaca,  65. 

polyglotta,  65. 

Ibis,  Glossy,  212. 

— ,  Red-cheeked,  211. 
Ibis  comatus,  211. 
lynx  torquilla,  129. 

Jackdaw,  83. 

Jacksnipe,  279. 

Jay,  Algerian  Black-headed,  82. 

,  West-European,  81. 

Kestrel,  Common,  195. 

,  Lesser,  196. 

Kingfisher,  130. 


Kite,  Black,  187. 
— ,  Red,  187. 
Kittiwake,  301. 
Knot,  281. 

Lammergeyer,  156. 
Landrail,  244. 
Lanius  algeriensis,  87. 

collurio,  88. 

—  meridionalis,  87. 

minor,  88. 

pomeranus,  89. 

Lanner,  192. 
Lapwing,  271. 

— ,  Black-bellied,  272. 
Lark,  Andalucian  Short- toed,  120. 

,  Desert  Curved-billed,  117. 

,  Dupont's,  117. 

,  Calandra,  121. 

— ,  Crested,  118. 
,  Horned  Desert,  121. 

— .  Short-toed,  120. 

,  Sky,  117. 

,  Small  Curved-billed,  117. 

— ,  Wood,  118. 
Larus  argentatus,  300. 

—  audouini,  299. 

cachinnans,  300. 

canus,  298. 

fuscus,  301. 

—  gelastes,  299. 

—  glaucus,  301. 

marinus,  301. 

melanocephalus,  297. 

minutus,  297. 

-  ridibundus,  297. 
Ligurinus  chloris,  98. 
Limosa  belgica,  288. 

lapponica,  288. 

Linnet,  Common,  104. 

,  Mountain,  105. 

Linota  canuabina,  104. 

—  flavirostris,  105. 

rufescens,  105. 

Locustella  luscinioides,  59. 
nasvia,  59. 


322 


INDEX  TO  BIRDS. 


Loxia  curvirostris,  106. 
Lusciniola  melanopogon,  61. 

Machetes  pugnax,  283. 
Magpie,  Common,  82. 
,  North-African,  82. 

— ,  Spanish  Azure-winged,  82. 
Mallard,  220. 
Mareca  penelope,  225. 
Marmaronetta  angustirostris,  223. 
Martin,  Crag,  96. 

— ,  House,  94. 

— ,  Sand,  95. 
Melierax  polyzonus,  168. 
Merganser,  Red-breasted,  231. 
Mergus  albellus,  232. 
merganser,  231. 

—  serrator,  231. 
Merlin,  194. 
Merops  apiaster,  131. 
Milvus  ictiuus,  187. 

• migrans,  187. 

Monticola  cyanus,  36. 

—  saxatilis,  38. 
Montifringilla  nivalis,  104. 
Moorhen,  247. 
Motacilla  alba,  111. 
borealis,  114. 

• cinereocapilla,  114. 

-  flava,  113. 

—  lugubris,  112. 

—  melanope,  112. 

—  raii,  114. 

Muscicapa  atricapilla,  92. 
collaris,  92. 

grisola,  91. 

Neophron  percn  op  torus,  152. 
Nightingale,  47. 
Nightjar,  125. 

— ,  Rufous-naped,  126. 
Nisaetus  fasciatus,  174. 

—  pennatus,  180. 
Nucifraga  caryocatactes,  80. 
Numenius  arquata,  290. 
hudsonicus,  291. 


Numeuius  phaeopus,  291. 

—  tenuirostris,  290. 
Nutcracker,  80. 
Nuthatch,  Common,  74. 
Nycticorax  griseus,  205. 

Oceanites  oceanicus,  310. 
OEdemia  fusca,  230. 

-  uigra,  229. 

CEdicnemus  scolopax,  262. 
Oriole,  Golden,  76. 
Oriolus  galbula,  76. 
Ortolan,  109. 
Osprey,  197. 
Otis  tarda,  253. 

—  tctrax,  257. 
Otocorys  bilopha,  121. 
Ouzel,  Ring,  36. 

— ,  Water,  76. 
Owl,  Barn,  137. 

— ,  Brown,  141. 

— ,  Eagle,  143. 

— ,  Little,  144. 

— ,  Long-eared,  139. 

— ,  Marsh,  140. 

— ,  Scops,  142. 

— ,  Short-eared,  139. 

— ,  Southern  Little,  145. 

— ,  Tawny,  141. 

— ,  White,  137. 

— ,  Wood,  141. 
Oystercatcher,  273. 

Pandion  haliaetus,  197. 
Pauurus  biarmicus,  70. 
Partridge,  Barbary,  239. 

,  French,  237. 

— ,  Western  Red-lcggcd, 
Parus  ater,  73. 
casrulcus,  72. 

—  cristatus,  73. 

—  ledouci,  73. 
major,  71. 

—  palustris,  73. 

—  teneriffae,  72. 
Passer  domesticus,  101. 


INDEX  TO  BIRDS. 


323 


Passer  hispaniolensis,  102. 

—  montanus,  102. 
Pastor  roseus,  79. 
Peewit,  271. 
Pernis  apivorus,  189. 
Petrel,  Fork-tailed,  310. 

— ,  Long-legged,  310. 

— ,  Storm,  309. 

— ,  Wilson's,  310. 
Petronia  stulta,  102. 
Phalacrocorax  carbo,  198. 
desmarestii,  198. 

—  graculus,  198. 
Phalarope,  Grey,  275. 
Phalaropus  fulicarius,  275. 
Phcenicopterus  roseus,  213. 
Phylloscopus  bonellii,  63. 

-  rufus,  63. 

—  sibilatrix,  63. 

—  trochilus,  64. 
Pica  raauritanica,  82. 

—  rustica,  82. 
Pigeon,  Wood,  232. 
Pintail,  222. 
Pipit,  Meadow,  115. 
,  Red-throated,  115. 

— ,  Richard's,  116. 

— ,  Rock,  115. 

— ,  Tawny,  116. 

— ,  Tree,  114. 

— ,  Water,  H6. 
Platalea  leucorodia,  210. 
Plectrophenax  nivalis,  111. 
Plegadis  falcinellus,  212. 
Plover,  Asiatic  Golden,  267. 

— ,  Golden,  266. 

— ,  Green,  271. 

— ,  Grey,  267. 

— ,  Kentish,  270. 

— ,  Little  Ringed,  269. 

— ,  Ringed,  268. 
Pochard,  Common,  226. 

— ,  Red-crested,  226. 

— ,  White-eyed,  227. 
Podicipes  auritas,  307. 
cristatus,  306. 


Podicipes  fluviatilis,  308. 
— •  griseigena,  307. 

—  nigricollis,  307. 
Porphyrio  caeruleus,  247. 
Porzana  bailloni,  245. 

—  maruetta,  244. 

—  parva,  245. 
Pratincola  rubetra,  43. 

—  rubicola,  43. 
Pratincole,  261. 
Procellaria  leucorrhoa  31 

pelagica,  309. 

Pterocles  arenarius,  237. 

senegallus,  237. 

Pteroclurus  alchata,  236. 

pyrenaicus,  236. 

Puffin,  305. 

Puffinus  anglorum,  309. 

—  gravis,  308. 
-  kuhli,  308. 

—  major,  308. 

—  obscurus,  309. 
—  yelkouan,  309. 

Pycnonotus  barbatus,  67. 
Pyrrhocorax  alpinus,  80. 

graculus,  80. 

Pyrrhula  erythrina,  106. 

Quail,  Andalucian  Bush,  241. 
— ,  Button,  244. 

,  Common,  240. 

-,  Three-toed,  241. 


Querquedula  circia,  225. 
•  crecca,  224. 

Rail,  Water,  246. 
Rallus  aquaticus,  246. 
Raven,  Common,  84. 

,  Tangier,  84. 

Razor-bill,  304. 
Recurvirostra  avocetta,  274. 
Redpole,  Lesser,  105. 
Redshank,  Common,  287. 

,  Dusky,  287. 

Redstart,  Black,  45. 
,  Common,  44. 


324 


INDEX  TO  BIEDS. 


Redstart,  Moussier's,  44. 
Redwing,  35. 
Reedling,  Bearded,  70. 
Reeve,  283. 
Regulus  cristatus,  66. 

ignicapillus,  66. 

Rissa  tridactyla,  301. 

Robin,  46. 

Roller,  130. 

Rook,  86. 

Ruff,  283. 

Ruticilla  cyanecula,  46. 

moussieri,  44. 

phoenicurus,  44. 

suecica,  46. 

—  titys,  45. 

-  wolfi,  46. 

Sanderling,  280. 

Sand-Grouse,  Black-bellied,  237. 

— ,  Pin-tailed,  236. 
Sandpiper,  Common,  285. 

— ,  Curlew,  280. 

,  Green,  284. 

,  Marsh,  286. 

,  Purple,  282. 

— ,  Terek,  286. 

— ,  Wood,  285. 
Saxicola  albicollis,  40. 

• leucura,  41. 

• cenanthe,  39. 

—  stapazina,  40. 
Scolopax  rusticula,  276. 
Scops  giu,  142. 
Scoter,  Common,  229. 

— ,  Velvet,  230. 
Serinus  hortulanus,  101. 
Shag,  198. 
Shearwater,  Cinereous,  308. 

— ,  Dusky,  309. 

— ,  Manx,  309. 
Sheld-duck,  219. 

,  Ruddy,  219. 

Shoveler,  222. 

Shrike,  Algerian  Grey,  87. 

,  Hooded,  90. 


Shrike,  Lesser  Grey,  88. 

— ,  Red-backed,  88. 

— ,  Spanish  Grey,  87. 

— ,  Woodchat,  89. 
Siskin,  100. 
Sitta  csesia,  74. 
Skua,  Arctic,  303. 

— ,  Buffon's,  303. 

— ,  Great,  302. 

— ,  Long-tailed,  303. 

— ,  Pomatorhine,  302. 

— ,  Richardson's,  303. 

— ,  Twist-tailed,  302. 
Smew,  232. 
Snipe,  Common,  278. 

— ,  Great,  277. 

— ,  Sabine's,  279. 

,  Solitary,  277. 

Sparrow,  Hedge,  49. 
,  House,  101. 

— ,  Rock,  102. 
,  Spanish,  102. 

— ,  Tree,  102. 
Sparrowhawk,  186. 
Spatula  clypeata,  222. 
Spoonbill,  White,  210. 
Squatarola  helvetica,  267. 
Starling,  Common,  77. 

,  Rose-coloured,  79. 

-,  Spotless,  78. 


Stercorarius  catarractes,  302. 

crepidatus,  303. 

parasiticus,  303. 

pomatorhinus,  302. 


Sterna  anglica,  293. 

cantiaca,  294. 

caspia,  294. 

-  fluviatilis,  296. 

macrura,  295. 

maxima,  294. 

media,  295. 

rainuta,  296. 


Stilt,  Black-winged,  274. 
Stint,  Little,  282. 

— ,  Lobed-footed,  275. 
,  Temminck's,  283. 


INDEX  TO  BIRDS. 


325 


Stonechat,  43.     , 
Stone-Curlew,  262. 
Stork,  Black,  210. 

,  White,  207. 

Strepsilas  interpres,  273. 
Strix  flammea,  137. 
Sturnus  unicolor,  78. 

vulgaris,  77. 

Sula  bassana,  199. 
Swallow,  Common,  93. 

,  Red-rumped,  94. 

Swan,  Mute,  216. 

,  Whooper,  215. 

Swift,  Alpine,  124. 

,  Common,  122. 

,  Mouse-coloured,  123. 

— ,  White-bellied,  124. 
Sylvia  atricapilla,  51.    . 

cinerea,  50. 

• conspicillata,  51. 

curruca,  50. 

• hortensis,  51. 

melanocephala,  54. 

orphea,  52. 

subalpina,  53. 

undata,  55. 

Syrnium  aluco,  141. 

Tadorna  casarca,  219. 

cornuta,  219. 

Tchagra,  90. 
Teal,  224. 

,  Summer,  225. 

Telephonus  erythropterus,  90. 
Tern,  AUied,  295. 

,  Arctic,  295. 

,  Black,  291. 

-,  Caspian,  294. 


,  Common,  296. 

,  Gull-billed,  293. 

,  Little,  296. 

,  Royal,  294. 

,  Sandwich,  294. 

,  Whiskered,  292. 

-,  White-winged  Black,  292. 


Thrush,  Blue  Rock,  36. 


Thrush,  Mistle,  34. 

— ,  Song,  34.    , 

— ,  White-backed  Rock,  38. 
Tichodroma  muraria,  75. 
Tit,  Algerian  Coal,  73. 
,  Blue,  72. 

— ,  Crested,  73. 

,  European  Coal,  73. 

,  Great,  71. 

,  Marsh,  73. 

,  Penduline,  74. 

,  Spanish  Long-tailed,  71. 

,  Ultramarine,  72. 

Totanus  calidris,  287. 

canescens,  286. 

cinereus,  286. 

fuscus,  287. 

glareola,  285. 

hypoleucus,  285. 

ochropus,  284. 

—  stagnatilis,  286. 
Tringa  alpina,  281. 

—  canutus,  281. 

—  minuta,  282. 
-  striata,  282, 

—  subarquata,  280. 

—  temmincki,  283. 
Troglodytes  parvulus,  75. 
Turdus  iliacus,  35. 

—  merula,  35. 
musicus,  34. 

—  pilaris,  35. 
—  torquatus,  36. 

—  viscivorus,  34. 
Turnix  sylvatica,  241. 
Turnstone,  273. 
Turtle-Dove,  Common,  234. 

,  Egyptian,  235. 

Turtur  communis,  234. 

senegalensis,  235. 

Twite,  105. 

Upupa  epops,  134. 
Uria  troile,  304. 

Yanellus  gregarius,  272. 


326 


INDEX  TO  BIRDS. 


Vanellus  vulgaris,  271. 
Vultur  monachus,  145. 
Vulture,  Bearded,  156. 

,  Black,  145. 

,  Cinereous,  145. 

,  Egyptian,  152. 

,  Griffon,  147. 

Wagtail,  Blue-headed  Yellow,  113. 

,  Grey,  112. 

• ,  Grey-headed,  114. 

,  Yellow,  114. 

— ,  Pied,  112. 

— ,  White,  111. 
Warbler,  Aquatic,  58. 
,  Black-headed,  54. 

— ,  Bonelli's  Willow,  63. 

,  Cetti's  Bush,  61. 

,  Chestnut-breasted,  53. 

,  Dartford,  55. 

,  Fantail,  69. 

,  Furze,  55. 

,  Garden,  51. 

,  Grass,  69. 

,  Grasshopper,  59. 

,  Great  Reed,  57. 

< ,  Marsh,  58. 

,  Moustached  Swamp,  61. 

,  Reed,  57. 

,  Savi's,  59. 

,  Sedge,  58. 

,  Short-winged  Yellow  Tree,  65. 


Warbler,  Spectacled,  51. 

,  Subalpine,  53. 

,  Western  Orphean,  52. 

,  Western  Pallid,  65. 

,  Western  Rufous,  56. 

,  Willow,  64. 

,  Wood,  63. 

Waterhen,  247. 

— ,  Purple,  247. 
Wheatear,  Black,  41. 

,  Black-eared,  40. 

,  Common,  39. 

,  Western  Black-throated,  40. 

Whimbrel,  291. 

,  American,  291. 

Whinchat,  43. 

White  throat,  Common,  50. 

— ,  Lesser,  50. 
Wigeon,  225. 
Woodcock,  276. 

Woodpecker,  Algerian  Green,  128. 
,  Algerian  Pied,  127. 

— ,  Great  Spotted,  126. 

— ,  Grey-headed  Green,  129. 

— ,  Middle  Spotted,  127. 

,  South  Spanish  Green,  128. 

Wood-Pigeon,  232. 
Wren,  75. 

— ,  Fire-crested,  66. 
— ,  Gold-crested,  66. 
Wryneck,  129. 


PRINTED  BY  TAYLOR  AND  FRANCIS,  KED  LION  COURT,   FLEET  STItEET. 


- 


KC\T. 


811998  r^F- 


UNIVERSITY  OF  CALIFORNIA  LIBRARY